From e2bcaee4f1c2293e8e569befca4eb88800cf47cd Mon Sep 17 00:00:00 2001 From: Hui Zhang Date: Wed, 3 Nov 2021 09:46:36 +0000 Subject: [PATCH] merge deepspeech, parakeet and text_processing into paddlespeech --- .mergify.yml | 6 +-- docs/source/asr/models_introduction.md | 10 ++-- docs/source/conf.py | 4 +- docs/source/index.rst | 4 +- docs/source/tts/advanced_usage.md | 8 ++-- docs/source/tts/quick_start.md | 12 ++--- examples/aishell/s0/path.sh | 2 +- examples/aishell/s1/path.sh | 2 +- examples/aishell3/tts3/path.sh | 2 +- examples/aishell3/vc0/README.md | 2 +- examples/aishell3/vc0/path.sh | 2 +- examples/callcenter/s1/path.sh | 2 +- examples/csmsc/tts2/path.sh | 2 +- examples/csmsc/tts3/path.sh | 2 +- examples/csmsc/voc1/path.sh | 2 +- examples/csmsc/voc3/path.sh | 2 +- examples/librispeech/s0/path.sh | 2 +- examples/librispeech/s1/path.sh | 2 +- examples/librispeech/s2/path.sh | 2 +- examples/ljspeech/tts0/path.sh | 2 +- examples/ljspeech/tts1/path.sh | 2 +- examples/ljspeech/tts3/path.sh | 2 +- examples/ljspeech/voc0/path.sh | 2 +- examples/ljspeech/voc1/path.sh | 2 +- .../other/1xt2x/src_deepspeech2x/__init__.py | 2 +- .../other/1xt2x/src_deepspeech2x/bin/test.py | 6 +-- .../models/ds2/deepspeech2.py | 10 ++-- .../1xt2x/src_deepspeech2x/models/ds2/rnn.py | 6 +-- .../1xt2x/src_deepspeech2x/test_model.py | 26 +++++------ examples/other/ge2e/path.sh | 2 +- examples/other/text_frontend/test_g2p.py | 4 +- examples/other/text_frontend/test_textnorm.py | 4 +- examples/ted_en_zh/t0/path.sh | 2 +- examples/timit/s1/path.sh | 2 +- examples/tiny/s0/path.sh | 2 +- examples/tiny/s1/path.sh | 2 +- examples/vctk/tts3/path.sh | 2 +- examples/vctk/voc1/path.sh | 2 +- hub/setup_hub.sh | 2 +- paddlespeech/__init__.py | 0 {deepspeech => paddlespeech/s2t}/__init__.py | 2 +- .../s2t}/decoders/README.md | 0 .../s2t}/decoders/__init__.py | 0 .../s2t}/decoders/beam_search/__init__.py | 0 .../decoders/beam_search/batch_beam_search.py | 0 .../s2t}/decoders/beam_search/beam_search.py | 2 +- .../s2t}/decoders/ctcdecoder/__init__.py | 0 .../ctcdecoder/decoders_deprecated.py | 0 .../decoders/ctcdecoder/scorer_deprecated.py | 0 .../s2t}/decoders/ctcdecoder/swig/.gitignore | 0 .../s2t}/decoders/ctcdecoder/swig/__init__.py | 0 .../swig/ctc_beam_search_decoder.cpp | 0 .../ctcdecoder/swig/ctc_beam_search_decoder.h | 0 .../ctcdecoder/swig/ctc_greedy_decoder.cpp | 0 .../ctcdecoder/swig/ctc_greedy_decoder.h | 0 .../ctcdecoder/swig/decoder_utils.cpp | 0 .../decoders/ctcdecoder/swig/decoder_utils.h | 0 .../s2t}/decoders/ctcdecoder/swig/decoders.i | 0 .../decoders/ctcdecoder/swig/path_trie.cpp | 0 .../s2t}/decoders/ctcdecoder/swig/path_trie.h | 0 .../s2t}/decoders/ctcdecoder/swig/scorer.cpp | 0 .../s2t}/decoders/ctcdecoder/swig/scorer.h | 0 .../s2t}/decoders/ctcdecoder/swig/setup.py | 0 .../s2t}/decoders/ctcdecoder/swig/setup.sh | 0 .../s2t}/decoders/ctcdecoder/swig_wrapper.py | 0 .../ctcdecoder/tests/test_decoders.py | 2 +- .../s2t}/decoders/recog.py | 10 ++-- .../s2t}/decoders/recog_bin.py | 2 +- .../s2t}/decoders/scorers/__init__.py | 0 .../s2t}/decoders/scorers/ctc.py | 2 +- .../s2t}/decoders/scorers/ctc_prefix_score.py | 0 .../s2t}/decoders/scorers/length_bonus.py | 0 .../s2t}/decoders/scorers/ngram.py | 0 .../s2t}/decoders/scorers/scorer_interface.py | 0 .../s2t}/decoders/utils.py | 2 +- .../s2t}/exps/__init__.py | 20 ++++---- .../s2t}/exps/deepspeech2/__init__.py | 0 .../exps/deepspeech2/bin/deploy/client.py | 2 +- .../exps/deepspeech2/bin/deploy/record.py | 0 .../exps/deepspeech2/bin/deploy/runtime.py | 20 ++++---- .../s2t}/exps/deepspeech2/bin/deploy/send.py | 2 +- .../exps/deepspeech2/bin/deploy/server.py | 20 ++++---- .../s2t}/exps/deepspeech2/bin/export.py | 8 ++-- .../s2t}/exps/deepspeech2/bin/test.py | 8 ++-- .../s2t}/exps/deepspeech2/bin/test_export.py | 8 ++-- .../s2t}/exps/deepspeech2/bin/test_hub.py | 22 ++++----- .../s2t}/exps/deepspeech2/bin/train.py | 8 ++-- .../s2t}/exps/deepspeech2/config.py | 12 ++--- .../s2t}/exps/deepspeech2/model.py | 38 +++++++-------- .../s2t}/exps/lm/transformer/__init__.py | 0 .../lm/transformer/bin/cacu_perplexity.py | 2 +- .../exps/lm/transformer/lm_cacu_perplexity.py | 8 ++-- .../s2t}/exps/u2/__init__.py | 0 .../s2t}/exps/u2/bin/alignment.py | 8 ++-- .../s2t}/exps/u2/bin/export.py | 8 ++-- .../s2t}/exps/u2/bin/test.py | 8 ++-- .../s2t}/exps/u2/bin/test_hub.py | 22 ++++----- .../s2t}/exps/u2/bin/train.py | 10 ++-- .../s2t}/exps/u2/config.py | 10 ++-- .../s2t}/exps/u2/model.py | 36 +++++++-------- .../s2t}/exps/u2/trainer.py | 34 +++++++------- .../s2t}/exps/u2_kaldi/__init__.py | 0 .../s2t}/exps/u2_kaldi/bin/recog.py | 2 +- .../s2t}/exps/u2_kaldi/bin/test.py | 10 ++-- .../s2t}/exps/u2_kaldi/bin/train.py | 10 ++-- .../s2t}/exps/u2_kaldi/model.py | 30 ++++++------ .../s2t}/exps/u2_st/__init__.py | 0 .../s2t}/exps/u2_st/bin/export.py | 8 ++-- .../s2t}/exps/u2_st/bin/test.py | 8 ++-- .../s2t}/exps/u2_st/bin/train.py | 8 ++-- .../s2t}/exps/u2_st/config.py | 10 ++-- .../s2t}/exps/u2_st/model.py | 34 +++++++------- .../s2t}/frontend/__init__.py | 0 .../s2t}/frontend/audio.py | 0 .../s2t}/frontend/augmentor/__init__.py | 0 .../s2t}/frontend/augmentor/augmentation.py | 22 ++++----- .../s2t}/frontend/augmentor/base.py | 0 .../frontend/augmentor/impulse_response.py | 6 +-- .../s2t}/frontend/augmentor/noise_perturb.py | 6 +-- .../online_bayesian_normalization.py | 2 +- .../s2t}/frontend/augmentor/resample.py | 2 +- .../s2t}/frontend/augmentor/shift_perturb.py | 2 +- .../s2t}/frontend/augmentor/spec_augment.py | 4 +- .../s2t}/frontend/augmentor/speed_perturb.py | 2 +- .../s2t}/frontend/augmentor/volume_perturb.py | 2 +- .../s2t}/frontend/featurizer/__init__.py | 0 .../frontend/featurizer/audio_featurizer.py | 0 .../frontend/featurizer/speech_featurizer.py | 4 +- .../frontend/featurizer/text_featurizer.py | 2 +- .../s2t}/frontend/normalizer.py | 8 ++-- .../s2t}/frontend/speech.py | 2 +- .../s2t}/frontend/utility.py | 2 +- .../s2t}/io/__init__.py | 0 .../s2t}/io/batchfy.py | 2 +- .../s2t}/io/collator.py | 20 ++++---- .../s2t}/io/converter.py | 4 +- .../s2t}/io/dataloader.py | 12 ++--- .../s2t}/io/dataset.py | 4 +- {deepspeech => paddlespeech/s2t}/io/reader.py | 4 +- .../s2t}/io/sampler.py | 2 +- .../s2t}/io/utility.py | 2 +- .../s2t}/models/__init__.py | 0 .../s2t}/models/asr_interface.py | 10 ++-- .../s2t}/models/ds2/__init__.py | 0 .../s2t}/models/ds2/conv.py | 6 +-- .../s2t}/models/ds2/deepspeech2.py | 12 ++--- .../s2t}/models/ds2/rnn.py | 6 +-- .../s2t}/models/ds2_online/__init__.py | 0 .../s2t}/models/ds2_online/conv.py | 2 +- .../s2t}/models/ds2_online/deepspeech2.py | 10 ++-- .../s2t}/models/lm/__init__.py | 0 .../s2t}/models/lm/transformer.py | 10 ++-- .../s2t}/models/lm_interface.py | 6 +-- .../s2t}/models/st_interface.py | 4 +- .../s2t}/models/u2/__init__.py | 0 .../s2t}/models/u2/u2.py | 46 +++++++++---------- .../s2t}/models/u2/updater.py | 12 ++--- .../s2t}/models/u2_st/__init__.py | 0 .../s2t}/models/u2_st/u2_st.py | 34 +++++++------- .../s2t}/modules/__init__.py | 0 .../s2t}/modules/activation.py | 2 +- .../s2t}/modules/attention.py | 2 +- .../s2t}/modules/cmvn.py | 2 +- .../s2t}/modules/conformer_convolution.py | 2 +- .../s2t}/modules/crf.py | 2 +- .../s2t}/modules/ctc.py | 12 ++--- .../s2t}/modules/decoder.py | 18 ++++---- .../s2t}/modules/decoder_layer.py | 2 +- .../s2t}/modules/embedding.py | 2 +- .../s2t}/modules/encoder.py | 34 +++++++------- .../s2t}/modules/encoder_layer.py | 2 +- .../s2t}/modules/loss.py | 2 +- .../s2t}/modules/mask.py | 2 +- .../s2t}/modules/positionwise_feed_forward.py | 2 +- .../s2t}/modules/subsampling.py | 4 +- .../s2t}/training/__init__.py | 0 .../s2t}/training/cli.py | 8 ++-- .../s2t}/training/extensions/__init__.py | 0 .../s2t}/training/extensions/evaluator.py | 2 +- .../s2t}/training/extensions/extension.py | 0 .../s2t}/training/extensions/plot.py | 0 .../s2t}/training/extensions/snapshot.py | 4 +- .../s2t}/training/extensions/visualizer.py | 0 .../s2t}/training/gradclip.py | 2 +- .../s2t}/training/optimizer.py | 8 ++-- .../s2t}/training/reporter.py | 0 .../s2t}/training/scheduler.py | 6 +-- .../s2t}/training/timer.py | 2 +- .../s2t}/training/trainer.py | 20 ++++---- .../s2t}/training/triggers/__init__.py | 0 .../triggers/compare_value_trigger.py | 0 .../training/triggers/interval_trigger.py | 0 .../s2t}/training/triggers/limit_trigger.py | 0 .../s2t}/training/triggers/time_trigger.py | 0 .../s2t}/training/triggers/utils.py | 0 .../s2t}/training/updaters/__init__.py | 0 .../training/updaters/standard_updater.py | 8 ++-- .../s2t}/training/updaters/trainer.py | 12 ++--- .../s2t}/training/updaters/updater.py | 2 +- .../s2t}/transform/__init__.py | 0 .../s2t}/transform/add_deltas.py | 0 .../s2t}/transform/channel_selector.py | 0 .../s2t}/transform/cmvn.py | 0 .../s2t}/transform/functional.py | 4 +- .../s2t}/transform/perturb.py | 2 +- .../s2t}/transform/spec_augment.py | 2 +- .../s2t}/transform/spectrogram.py | 0 .../s2t}/transform/transform_interface.py | 0 .../s2t}/transform/transformation.py | 42 ++++++++--------- .../s2t}/transform/wpe.py | 0 .../s2t}/utils/__init__.py | 0 .../s2t}/utils/asr_utils.py | 0 .../s2t}/utils/bleu_score.py | 0 .../s2t}/utils/check_kwargs.py | 0 .../s2t}/utils/checkpoint.py | 4 +- .../s2t}/utils/cli_readers.py | 2 +- .../s2t}/utils/cli_utils.py | 0 .../s2t}/utils/cli_writers.py | 4 +- .../s2t}/utils/ctc_utils.py | 6 +-- .../s2t}/utils/dynamic_import.py | 8 ++-- .../s2t}/utils/error_rate.py | 0 .../s2t}/utils/layer_tools.py | 0 {deepspeech => paddlespeech/s2t}/utils/log.py | 0 .../s2t}/utils/mp_tools.py | 0 .../s2t}/utils/profiler.py | 2 +- .../s2t}/utils/socket_server.py | 2 +- .../s2t}/utils/spec_augment.py | 0 .../s2t}/utils/tensor_utils.py | 2 +- .../s2t}/utils/text_grid.py | 0 .../s2t}/utils/utility.py | 2 +- {parakeet => paddlespeech/t2s}/__init__.py | 0 .../t2s}/audio/__init__.py | 0 {parakeet => paddlespeech/t2s}/audio/audio.py | 0 .../t2s}/audio/spec_normalizer.py | 0 .../t2s}/data/__init__.py | 2 +- {parakeet => paddlespeech/t2s}/data/batch.py | 0 .../t2s}/data/dataset.py | 0 .../t2s}/data/get_feats.py | 0 .../t2s}/datasets/__init__.py | 0 .../t2s}/datasets/am_batch_fn.py | 2 +- .../t2s}/datasets/common.py | 0 .../t2s}/datasets/data_table.py | 0 .../t2s}/datasets/ljspeech.py | 0 .../t2s}/datasets/preprocess_utils.py | 0 .../t2s}/datasets/vocoder_batch_fn.py | 0 .../t2s}/exps/__init__.py | 0 .../t2s}/exps/fastspeech2/__init__.py | 0 .../t2s}/exps/fastspeech2/inference.py | 2 +- .../fastspeech2/multi_spk_synthesize_e2e.py | 12 ++--- .../multi_spk_synthesize_e2e_en.py | 12 ++--- .../t2s}/exps/fastspeech2/normalize.py | 2 +- .../t2s}/exps/fastspeech2/preprocess.py | 16 +++---- .../t2s}/exps/fastspeech2/synthesize.py | 12 ++--- .../t2s}/exps/fastspeech2/synthesize_e2e.py | 12 ++--- .../exps/fastspeech2/synthesize_e2e_en.py | 12 ++--- .../exps/fastspeech2/synthesize_e2e_melgan.py | 12 ++--- .../t2s}/exps/fastspeech2/train.py | 22 ++++----- .../t2s}/exps/gan_vocoder/README.md | 0 .../t2s}/exps/gan_vocoder/__init__.py | 0 .../gan_vocoder/multi_band_melgan/__init__.py | 0 .../multi_band_melgan/synthesize.py | 4 +- .../gan_vocoder/multi_band_melgan/train.py | 28 +++++------ .../t2s}/exps/gan_vocoder/normalize.py | 2 +- .../gan_vocoder/parallelwave_gan/__init__.py | 0 .../parallelwave_gan/synthesize.py | 4 +- .../parallelwave_gan/synthesize_from_wav.py | 8 ++-- .../gan_vocoder/parallelwave_gan/train.py | 22 ++++----- .../t2s}/exps/gan_vocoder/preprocess.py | 6 +-- .../t2s}/exps/ge2e/__init__.py | 0 .../t2s}/exps/ge2e/audio_processor.py | 0 .../t2s}/exps/ge2e/config.py | 0 .../t2s}/exps/ge2e/dataset_processors.py | 2 +- .../t2s}/exps/ge2e/inference.py | 6 +-- .../t2s}/exps/ge2e/preprocess.py | 12 ++--- .../t2s}/exps/ge2e/random_cycle.py | 0 .../exps/ge2e/speaker_verification_dataset.py | 2 +- .../t2s}/exps/ge2e/train.py | 14 +++--- .../t2s}/exps/sentences.txt | 0 .../t2s}/exps/sentences_en.txt | 0 .../t2s}/exps/speedyspeech/__init__.py | 0 .../t2s}/exps/speedyspeech/inference.py | 2 +- .../t2s}/exps/speedyspeech/normalize.py | 2 +- .../t2s}/exps/speedyspeech/preprocess.py | 10 ++-- .../t2s}/exps/speedyspeech/synthesize.py | 12 ++--- .../t2s}/exps/speedyspeech/synthesize_e2e.py | 12 ++--- .../t2s}/exps/speedyspeech/train.py | 20 ++++---- .../t2s}/exps/tacotron2/__init__.py | 0 .../t2s}/exps/tacotron2/config.py | 0 .../t2s}/exps/tacotron2/ljspeech.py | 4 +- .../t2s}/exps/tacotron2/preprocess.py | 10 ++-- .../t2s}/exps/tacotron2/synthesize.ipynb | 10 ++-- .../t2s}/exps/tacotron2/synthesize.py | 8 ++-- .../t2s}/exps/tacotron2/train.py | 20 ++++---- .../t2s}/exps/transformer_tts/__init__.py | 0 .../t2s}/exps/transformer_tts/normalize.py | 2 +- .../t2s}/exps/transformer_tts/preprocess.py | 4 +- .../t2s}/exps/transformer_tts/synthesize.py | 12 ++--- .../exps/transformer_tts/synthesize_e2e.py | 12 ++--- .../t2s}/exps/transformer_tts/train.py | 20 ++++---- .../t2s}/exps/voice_cloning/__init__.py | 0 .../voice_cloning/tacotron2_ge2e/__init__.py | 0 .../voice_cloning/tacotron2_ge2e/aishell3.py | 10 ++-- .../tacotron2_ge2e/chinese_g2p.py | 2 +- .../voice_cloning/tacotron2_ge2e/config.py | 0 .../tacotron2_ge2e/extract_mel.py | 8 ++-- .../voice_cloning/tacotron2_ge2e/lexicon.txt | 0 .../preprocess_transcription.py | 0 .../tacotron2_ge2e/process_wav.py | 0 .../voice_cloning/tacotron2_ge2e/train.py | 20 ++++---- .../tacotron2_ge2e/voice_cloning.py | 16 +++---- .../t2s}/exps/waveflow/__init__.py | 0 .../t2s}/exps/waveflow/config.py | 0 .../t2s}/exps/waveflow/ljspeech.py | 4 +- .../t2s}/exps/waveflow/preprocess.py | 6 +-- .../t2s}/exps/waveflow/synthesize.py | 6 +-- .../t2s}/exps/waveflow/train.py | 20 ++++---- .../t2s}/frontend/__init__.py | 0 .../t2s}/frontend/arpabet.py | 4 +- .../t2s}/frontend/generate_lexicon.py | 0 .../t2s/frontend/normalizer/__init__.py | 15 ++++++ .../t2s}/frontend/normalizer/abbrrviation.py | 0 .../t2s}/frontend/normalizer/acronyms.py | 0 .../t2s}/frontend/normalizer/normalizer.py | 2 +- .../t2s}/frontend/normalizer/numbers.py | 0 .../t2s}/frontend/normalizer/width.py | 0 .../t2s}/frontend/phonectic.py | 6 +-- .../t2s}/frontend/pinyin.py | 4 +- .../t2s}/frontend/punctuation.py | 0 .../t2s}/frontend/tone_sandhi.py | 0 .../t2s}/frontend/vocab.py | 0 .../t2s}/frontend/zh_frontend.py | 6 +-- .../t2s}/frontend/zh_normalization/README.md | 0 .../t2s/frontend/zh_normalization/__init__.py | 14 ++++++ .../frontend/zh_normalization/char_convert.py | 0 .../frontend/zh_normalization/chronology.py | 0 .../frontend/zh_normalization/constants.py | 0 .../t2s}/frontend/zh_normalization/num.py | 0 .../frontend/zh_normalization/phonecode.py | 0 .../frontend/zh_normalization/quantifier.py | 0 .../zh_normalization/text_normlization.py | 0 .../t2s}/models/__init__.py | 0 .../t2s}/models/fastspeech2/__init__.py | 0 .../t2s}/models/fastspeech2/fastspeech2.py | 22 ++++----- .../models/fastspeech2/fastspeech2_updater.py | 8 ++-- .../t2s}/models/lstm_speaker_encoder.py | 0 .../t2s}/models/melgan/__init__.py | 0 .../t2s}/models/melgan/melgan.py | 10 ++-- .../melgan/multi_band_melgan_updater.py | 8 ++-- .../t2s}/models/parallel_wavegan/__init__.py | 0 .../parallel_wavegan/parallel_wavegan.py | 0 .../parallel_wavegan_updater.py | 8 ++-- .../t2s}/models/speedyspeech/__init__.py | 0 .../t2s}/models/speedyspeech/speedyspeech.py | 4 +- .../speedyspeech/speedyspeech_updater.py | 12 ++--- .../t2s}/models/tacotron2.py | 8 ++-- .../t2s}/models/transformer_tts/__init__.py | 0 .../models/transformer_tts/transformer_tts.py | 26 +++++------ .../transformer_tts_updater.py | 10 ++-- .../t2s}/models/waveflow.py | 4 +- .../t2s}/modules/__init__.py | 0 .../t2s}/modules/adversarial_loss.py | 0 .../t2s}/modules/attention.py | 0 .../t2s}/modules/audio.py | 0 .../t2s}/modules/causal_conv.py | 0 .../t2s}/modules/conv.py | 0 .../t2s}/modules/expansion.py | 0 .../modules/fastspeech2_predictor/__init__.py | 0 .../duration_predictor.py | 4 +- .../fastspeech2_predictor/length_regulator.py | 0 .../variance_predictor.py | 4 +- .../fastspeech2_transformer/__init__.py | 0 .../fastspeech2_transformer/attention.py | 2 +- .../fastspeech2_transformer/decoder.py | 16 +++---- .../fastspeech2_transformer/decoder_layer.py | 2 +- .../fastspeech2_transformer/embedding.py | 0 .../fastspeech2_transformer/encoder.py | 14 +++--- .../fastspeech2_transformer/encoder_layer.py | 0 .../fastspeech2_transformer/lightconv.py | 4 +- .../modules/fastspeech2_transformer/mask.py | 0 .../multi_layer_conv.py | 0 .../positionwise_feed_forward.py | 0 .../modules/fastspeech2_transformer/repeat.py | 0 .../t2s}/modules/geometry.py | 0 {parakeet => paddlespeech/t2s}/modules/glu.py | 0 .../t2s}/modules/layer_norm.py | 0 .../t2s}/modules/losses.py | 0 .../t2s}/modules/masked_fill.py | 0 .../t2s}/modules/masking.py | 0 .../t2s}/modules/nets_utils.py | 0 .../t2s}/modules/normalizer.py | 0 .../t2s}/modules/positional_encoding.py | 0 .../t2s}/modules/pqmf.py | 0 .../t2s}/modules/residual_stack.py | 2 +- .../t2s}/modules/ssim.py | 0 .../t2s}/modules/stft_loss.py | 0 .../t2s}/modules/style_encoder.py | 2 +- .../t2s}/modules/tacotron2/__init__.py | 0 .../t2s}/modules/tacotron2/decoder.py | 0 .../t2s}/modules/tacotron2/encoder.py | 0 .../t2s}/modules/transformer.py | 2 +- .../t2s}/training/__init__.py | 0 .../t2s}/training/cli.py | 8 ++-- .../t2s}/training/default_config.py | 0 .../t2s}/training/experiment.py | 4 +- .../t2s}/training/extension.py | 0 .../t2s}/training/extensions/__init__.py | 0 .../t2s}/training/extensions/evaluator.py | 8 ++-- .../t2s}/training/extensions/snapshot.py | 6 +-- .../t2s}/training/extensions/visualizer.py | 4 +- .../t2s}/training/optimizer.py | 0 .../t2s}/training/reporter.py | 0 .../t2s}/training/seeding.py | 0 .../t2s}/training/trainer.py | 14 +++--- .../t2s}/training/trigger.py | 2 +- .../t2s}/training/triggers/__init__.py | 0 .../training/triggers/interval_trigger.py | 0 .../t2s}/training/triggers/limit_trigger.py | 0 .../t2s}/training/triggers/time_trigger.py | 0 .../t2s}/training/updater.py | 0 .../t2s}/training/updaters/__init__.py | 0 .../training/updaters/standard_updater.py | 6 +-- .../t2s}/utils/__init__.py | 0 .../t2s}/utils/checkpoint.py | 2 +- .../t2s}/utils/display.py | 0 .../t2s}/utils/error_rate.py | 0 .../t2s}/utils/h5_utils.py | 0 .../t2s}/utils/internals.py | 0 .../t2s}/utils/layer_tools.py | 0 .../t2s}/utils/mp_tools.py | 0 .../t2s}/utils/profile.py | 0 .../t2s}/utils/profiler.py | 0 .../t2s}/utils/scheduler.py | 0 .../t2s}/utils/timeline.py | 0 .../text}/.gitignore | 0 .../text}/README.md | 0 .../punctuation_restoration/chinese/README.md | 0 .../chinese/conf/blstm.yaml | 0 .../chinese/conf/data_conf/chinese.yaml | 0 .../chinese/conf/train_conf/bertBLSTM_zh.yaml | 0 .../conf/train_conf/bertLinear_zh.yaml | 0 .../chinese/local/avg.sh | 0 .../chinese/local/data.sh | 0 .../chinese/local/test.sh | 0 .../chinese/local/train.sh | 0 .../punctuation_restoration/chinese/path.sh | 0 .../punctuation_restoration/chinese/run.sh | 0 .../punctuation_restoration/english/README.md | 0 .../english/conf/data_conf/english.yaml | 0 .../conf/train_conf/bertBLSTM_base_en.yaml | 0 .../conf/train_conf/bertLinear_en.yaml | 0 .../english/local/avg.sh | 0 .../english/local/data.sh | 0 .../english/local/test.sh | 0 .../english/local/train.sh | 0 .../punctuation_restoration/english/path.sh | 0 .../punctuation_restoration/english/run.sh | 0 .../text}/requirements.txt | 0 .../punctuation_restoration/bin/avg_model.py | 0 .../punctuation_restoration/bin/pre_data.py | 0 .../punctuation_restoration/bin/test.py | 0 .../punctuation_restoration/bin/train.py | 0 .../punctuation_restoration/io/__init__.py | 0 .../punctuation_restoration/io/collator.py | 0 .../punctuation_restoration/io/common.py | 0 .../punctuation_restoration/io/dataset.py | 0 .../model/BertBLSTM.py | 0 .../model/BertLinear.py | 0 .../punctuation_restoration/model/blstm.py | 0 .../punctuation_restoration/model/lstm.py | 0 .../modules/__init__.py | 0 .../modules/activation.py | 0 .../modules/attention.py | 0 .../punctuation_restoration/modules/crf.py | 0 .../training/__init__.py | 0 .../punctuation_restoration/training/loss.py | 0 .../training/trainer.py | 0 .../punctuation_restoration/utils/__init__.py | 0 .../utils/checkpoint.py | 0 .../utils/default_parser.py | 8 ++-- .../utils/layer_tools.py | 0 .../punctuation_restoration/utils/mp_tools.py | 0 .../utils/punct_pre.py | 0 .../punctuation_restoration/utils/utility.py | 0 parakeet/frontend/normalizer/__init__.py | 15 ------ .../frontend/zh_normalization/__init__.py | 14 ------ setup.py | 2 +- setup.sh | 2 +- tests/benchmark/pwgan/README.md | 2 +- tests/benchmark/pwgan/run_all.sh | 8 ++-- tests/benchmark/pwgan/run_benchmark.sh | 4 +- .../ds2/ds2_params_lite_train_infer.txt | 6 +-- .../ds2/ds2_params_whole_train_infer.txt | 6 +-- .../speedyspeech_params_lite_multi_gpu.txt | 6 +-- .../speedyspeech_params_lite_single_gpu.txt | 6 +-- .../speedyspeech_params_whole_multi_gpu.txt | 6 +-- .../speedyspeech_params_whole_single_gpu.txt | 6 +-- tests/unit/asr/deepspeech2_model_test.py | 2 +- .../unit/asr/deepspeech2_online_model_test.py | 2 +- tests/unit/asr/error_rate_test.py | 2 +- tests/unit/asr/mask_test.py | 4 +- tests/unit/asr/u2_model_test.py | 4 +- tests/unit/tts/test_data_table.py | 2 +- tests/unit/tts/test_expansion.py | 2 +- tests/unit/tts/test_pwg.py | 14 +++--- tests/unit/tts/test_reporter.py | 6 +-- tests/unit/tts/test_snapshot.py | 6 +-- tests/unit/tts/test_stft.py | 4 +- utils/apply-cmvn.py | 10 ++-- utils/build_vocab.py | 16 +++---- utils/compute-cmvn-stats.py | 10 ++-- utils/compute_mean_std.py | 10 ++-- utils/compute_statistics.py | 2 +- utils/copy-feats.py | 10 ++-- utils/dump_manifest.py | 2 +- utils/feat-to-shape.py | 8 ++-- utils/format_data.py | 10 ++-- utils/format_triplet_data.py | 10 ++-- utils/merge_scp2json.py | 2 +- 518 files changed, 1058 insertions(+), 1058 deletions(-) create mode 100644 paddlespeech/__init__.py rename {deepspeech => paddlespeech/s2t}/__init__.py (99%) rename {deepspeech => paddlespeech/s2t}/decoders/README.md (100%) rename {deepspeech => paddlespeech/s2t}/decoders/__init__.py (100%) rename {deepspeech => paddlespeech/s2t}/decoders/beam_search/__init__.py (100%) rename {deepspeech => paddlespeech/s2t}/decoders/beam_search/batch_beam_search.py (100%) rename {deepspeech => paddlespeech/s2t}/decoders/beam_search/beam_search.py (99%) rename {deepspeech => paddlespeech/s2t}/decoders/ctcdecoder/__init__.py (100%) rename {deepspeech => paddlespeech/s2t}/decoders/ctcdecoder/decoders_deprecated.py (100%) rename {deepspeech => paddlespeech/s2t}/decoders/ctcdecoder/scorer_deprecated.py (100%) rename {deepspeech => paddlespeech/s2t}/decoders/ctcdecoder/swig/.gitignore (100%) rename {deepspeech => paddlespeech/s2t}/decoders/ctcdecoder/swig/__init__.py (100%) rename {deepspeech => paddlespeech/s2t}/decoders/ctcdecoder/swig/ctc_beam_search_decoder.cpp (100%) rename {deepspeech => paddlespeech/s2t}/decoders/ctcdecoder/swig/ctc_beam_search_decoder.h (100%) rename {deepspeech => paddlespeech/s2t}/decoders/ctcdecoder/swig/ctc_greedy_decoder.cpp (100%) rename {deepspeech => paddlespeech/s2t}/decoders/ctcdecoder/swig/ctc_greedy_decoder.h (100%) rename {deepspeech => paddlespeech/s2t}/decoders/ctcdecoder/swig/decoder_utils.cpp (100%) rename {deepspeech => paddlespeech/s2t}/decoders/ctcdecoder/swig/decoder_utils.h (100%) rename {deepspeech => paddlespeech/s2t}/decoders/ctcdecoder/swig/decoders.i (100%) rename {deepspeech => paddlespeech/s2t}/decoders/ctcdecoder/swig/path_trie.cpp (100%) rename {deepspeech => paddlespeech/s2t}/decoders/ctcdecoder/swig/path_trie.h (100%) rename {deepspeech => paddlespeech/s2t}/decoders/ctcdecoder/swig/scorer.cpp (100%) rename {deepspeech => paddlespeech/s2t}/decoders/ctcdecoder/swig/scorer.h (100%) rename {deepspeech => paddlespeech/s2t}/decoders/ctcdecoder/swig/setup.py (100%) rename {deepspeech => paddlespeech/s2t}/decoders/ctcdecoder/swig/setup.sh (100%) rename {deepspeech => paddlespeech/s2t}/decoders/ctcdecoder/swig_wrapper.py (100%) rename {deepspeech => paddlespeech/s2t}/decoders/ctcdecoder/tests/test_decoders.py (98%) rename {deepspeech => paddlespeech/s2t}/decoders/recog.py (95%) rename {deepspeech => paddlespeech/s2t}/decoders/recog_bin.py (99%) rename {deepspeech => paddlespeech/s2t}/decoders/scorers/__init__.py (100%) rename {deepspeech => paddlespeech/s2t}/decoders/scorers/ctc.py (98%) rename {deepspeech => paddlespeech/s2t}/decoders/scorers/ctc_prefix_score.py (100%) rename {deepspeech => paddlespeech/s2t}/decoders/scorers/length_bonus.py (100%) rename {deepspeech => paddlespeech/s2t}/decoders/scorers/ngram.py (100%) rename {deepspeech => paddlespeech/s2t}/decoders/scorers/scorer_interface.py (100%) rename {deepspeech => paddlespeech/s2t}/decoders/utils.py (98%) rename {deepspeech => paddlespeech/s2t}/exps/__init__.py (69%) rename {deepspeech => paddlespeech/s2t}/exps/deepspeech2/__init__.py (100%) rename {deepspeech => paddlespeech/s2t}/exps/deepspeech2/bin/deploy/client.py (97%) rename {deepspeech => paddlespeech/s2t}/exps/deepspeech2/bin/deploy/record.py (100%) rename {deepspeech => paddlespeech/s2t}/exps/deepspeech2/bin/deploy/runtime.py (90%) rename {deepspeech => paddlespeech/s2t}/exps/deepspeech2/bin/deploy/send.py (96%) rename {deepspeech => paddlespeech/s2t}/exps/deepspeech2/bin/deploy/server.py (87%) rename {deepspeech => paddlespeech/s2t}/exps/deepspeech2/bin/export.py (85%) rename {deepspeech => paddlespeech/s2t}/exps/deepspeech2/bin/test.py (85%) rename {deepspeech => paddlespeech/s2t}/exps/deepspeech2/bin/test_export.py (85%) rename {deepspeech => paddlespeech/s2t}/exps/deepspeech2/bin/test_hub.py (90%) rename {deepspeech => paddlespeech/s2t}/exps/deepspeech2/bin/train.py (85%) rename {deepspeech => paddlespeech/s2t}/exps/deepspeech2/config.py (77%) rename {deepspeech => paddlespeech/s2t}/exps/deepspeech2/model.py (95%) rename {deepspeech => paddlespeech/s2t}/exps/lm/transformer/__init__.py (100%) rename {deepspeech => paddlespeech/s2t}/exps/lm/transformer/bin/cacu_perplexity.py (96%) rename {deepspeech => paddlespeech/s2t}/exps/lm/transformer/lm_cacu_perplexity.py (94%) rename {deepspeech => paddlespeech/s2t}/exps/u2/__init__.py (100%) rename {deepspeech => paddlespeech/s2t}/exps/u2/bin/alignment.py (85%) rename {deepspeech => paddlespeech/s2t}/exps/u2/bin/export.py (85%) rename {deepspeech => paddlespeech/s2t}/exps/u2/bin/test.py (86%) rename {deepspeech => paddlespeech/s2t}/exps/u2/bin/test_hub.py (90%) rename {deepspeech => paddlespeech/s2t}/exps/u2/bin/train.py (83%) rename {deepspeech => paddlespeech/s2t}/exps/u2/config.py (80%) rename {deepspeech => paddlespeech/s2t}/exps/u2/model.py (96%) rename {deepspeech => paddlespeech/s2t}/exps/u2/trainer.py (88%) rename {deepspeech => paddlespeech/s2t}/exps/u2_kaldi/__init__.py (100%) rename {deepspeech => paddlespeech/s2t}/exps/u2_kaldi/bin/recog.py (92%) rename {deepspeech => paddlespeech/s2t}/exps/u2_kaldi/bin/test.py (88%) rename {deepspeech => paddlespeech/s2t}/exps/u2_kaldi/bin/train.py (85%) rename {deepspeech => paddlespeech/s2t}/exps/u2_kaldi/model.py (96%) rename {deepspeech => paddlespeech/s2t}/exps/u2_st/__init__.py (100%) rename {deepspeech => paddlespeech/s2t}/exps/u2_st/bin/export.py (84%) rename {deepspeech => paddlespeech/s2t}/exps/u2_st/bin/test.py (86%) rename {deepspeech => paddlespeech/s2t}/exps/u2_st/bin/train.py (86%) rename {deepspeech => paddlespeech/s2t}/exps/u2_st/config.py (80%) rename {deepspeech => paddlespeech/s2t}/exps/u2_st/model.py (96%) rename {deepspeech => paddlespeech/s2t}/frontend/__init__.py (100%) rename {deepspeech => paddlespeech/s2t}/frontend/audio.py (100%) rename {deepspeech => paddlespeech/s2t}/frontend/augmentor/__init__.py (100%) rename {deepspeech => paddlespeech/s2t}/frontend/augmentor/augmentation.py (89%) rename {deepspeech => paddlespeech/s2t}/frontend/augmentor/base.py (100%) rename {deepspeech => paddlespeech/s2t}/frontend/augmentor/impulse_response.py (90%) rename {deepspeech => paddlespeech/s2t}/frontend/augmentor/noise_perturb.py (93%) rename {deepspeech => paddlespeech/s2t}/frontend/augmentor/online_bayesian_normalization.py (97%) rename {deepspeech => paddlespeech/s2t}/frontend/augmentor/resample.py (95%) rename {deepspeech => paddlespeech/s2t}/frontend/augmentor/shift_perturb.py (96%) rename {deepspeech => paddlespeech/s2t}/frontend/augmentor/spec_augment.py (98%) rename {deepspeech => paddlespeech/s2t}/frontend/augmentor/speed_perturb.py (98%) rename {deepspeech => paddlespeech/s2t}/frontend/augmentor/volume_perturb.py (96%) rename {deepspeech => paddlespeech/s2t}/frontend/featurizer/__init__.py (100%) rename {deepspeech => paddlespeech/s2t}/frontend/featurizer/audio_featurizer.py (100%) rename {deepspeech => paddlespeech/s2t}/frontend/featurizer/speech_featurizer.py (95%) rename {deepspeech => paddlespeech/s2t}/frontend/featurizer/text_featurizer.py (99%) rename {deepspeech => paddlespeech/s2t}/frontend/normalizer.py (96%) rename {deepspeech => paddlespeech/s2t}/frontend/speech.py (99%) rename {deepspeech => paddlespeech/s2t}/frontend/utility.py (99%) rename {deepspeech => paddlespeech/s2t}/io/__init__.py (100%) rename {deepspeech => paddlespeech/s2t}/io/batchfy.py (99%) rename {deepspeech => paddlespeech/s2t}/io/collator.py (96%) rename {deepspeech => paddlespeech/s2t}/io/converter.py (96%) rename {deepspeech => paddlespeech/s2t}/io/dataloader.py (94%) rename {deepspeech => paddlespeech/s2t}/io/dataset.py (99%) rename {deepspeech => paddlespeech/s2t}/io/reader.py (99%) rename {deepspeech => paddlespeech/s2t}/io/sampler.py (99%) rename {deepspeech => paddlespeech/s2t}/io/utility.py (98%) rename {deepspeech => paddlespeech/s2t}/models/__init__.py (100%) rename {deepspeech => paddlespeech/s2t}/models/asr_interface.py (94%) rename {deepspeech => paddlespeech/s2t}/models/ds2/__init__.py (100%) rename {deepspeech => paddlespeech/s2t}/models/ds2/conv.py (97%) rename {deepspeech => paddlespeech/s2t}/models/ds2/deepspeech2.py (97%) rename {deepspeech => paddlespeech/s2t}/models/ds2/rnn.py (98%) rename {deepspeech => paddlespeech/s2t}/models/ds2_online/__init__.py (100%) rename {deepspeech => paddlespeech/s2t}/models/ds2_online/conv.py (95%) rename {deepspeech => paddlespeech/s2t}/models/ds2_online/deepspeech2.py (98%) rename {deepspeech => paddlespeech/s2t}/models/lm/__init__.py (100%) rename {deepspeech => paddlespeech/s2t}/models/lm/transformer.py (96%) rename {deepspeech => paddlespeech/s2t}/models/lm_interface.py (91%) rename {deepspeech => paddlespeech/s2t}/models/st_interface.py (95%) rename {deepspeech => paddlespeech/s2t}/models/u2/__init__.py (100%) rename {deepspeech => paddlespeech/s2t}/models/u2/u2.py (96%) rename {deepspeech => paddlespeech/s2t}/models/u2/updater.py (93%) rename {deepspeech => paddlespeech/s2t}/models/u2_st/__init__.py (100%) rename {deepspeech => paddlespeech/s2t}/models/u2_st/u2_st.py (96%) rename {deepspeech => paddlespeech/s2t}/modules/__init__.py (100%) rename {deepspeech => paddlespeech/s2t}/modules/activation.py (99%) rename {deepspeech => paddlespeech/s2t}/modules/attention.py (99%) rename {deepspeech => paddlespeech/s2t}/modules/cmvn.py (97%) rename {deepspeech => paddlespeech/s2t}/modules/conformer_convolution.py (99%) rename {deepspeech => paddlespeech/s2t}/modules/crf.py (99%) rename {deepspeech => paddlespeech/s2t}/modules/ctc.py (96%) rename {deepspeech => paddlespeech/s2t}/modules/decoder.py (94%) rename {deepspeech => paddlespeech/s2t}/modules/decoder_layer.py (99%) rename {deepspeech => paddlespeech/s2t}/modules/embedding.py (99%) rename {deepspeech => paddlespeech/s2t}/modules/encoder.py (94%) rename {deepspeech => paddlespeech/s2t}/modules/encoder_layer.py (99%) rename {deepspeech => paddlespeech/s2t}/modules/loss.py (99%) rename {deepspeech => paddlespeech/s2t}/modules/mask.py (99%) rename {deepspeech => paddlespeech/s2t}/modules/positionwise_feed_forward.py (97%) rename {deepspeech => paddlespeech/s2t}/modules/subsampling.py (98%) rename {deepspeech => paddlespeech/s2t}/training/__init__.py (100%) rename {deepspeech => paddlespeech/s2t}/training/cli.py (95%) rename {deepspeech => paddlespeech/s2t}/training/extensions/__init__.py (100%) rename {deepspeech => paddlespeech/s2t}/training/extensions/evaluator.py (98%) rename {deepspeech => paddlespeech/s2t}/training/extensions/extension.py (100%) rename {deepspeech => paddlespeech/s2t}/training/extensions/plot.py (100%) rename {deepspeech => paddlespeech/s2t}/training/extensions/snapshot.py (97%) rename {deepspeech => paddlespeech/s2t}/training/extensions/visualizer.py (100%) rename {deepspeech => paddlespeech/s2t}/training/gradclip.py (98%) rename {deepspeech => paddlespeech/s2t}/training/optimizer.py (93%) rename {deepspeech => paddlespeech/s2t}/training/reporter.py (100%) rename {deepspeech => paddlespeech/s2t}/training/scheduler.py (95%) rename {deepspeech => paddlespeech/s2t}/training/timer.py (97%) rename {deepspeech => paddlespeech/s2t}/training/trainer.py (97%) rename {deepspeech => paddlespeech/s2t}/training/triggers/__init__.py (100%) rename {deepspeech => paddlespeech/s2t}/training/triggers/compare_value_trigger.py (100%) rename {deepspeech => paddlespeech/s2t}/training/triggers/interval_trigger.py (100%) rename {deepspeech => paddlespeech/s2t}/training/triggers/limit_trigger.py (100%) rename {deepspeech => paddlespeech/s2t}/training/triggers/time_trigger.py (100%) rename {deepspeech => paddlespeech/s2t}/training/triggers/utils.py (100%) rename {deepspeech => paddlespeech/s2t}/training/updaters/__init__.py (100%) rename {deepspeech => paddlespeech/s2t}/training/updaters/standard_updater.py (97%) rename {deepspeech => paddlespeech/s2t}/training/updaters/trainer.py (94%) rename {deepspeech => paddlespeech/s2t}/training/updaters/updater.py (98%) rename {deepspeech => paddlespeech/s2t}/transform/__init__.py (100%) rename {deepspeech => paddlespeech/s2t}/transform/add_deltas.py (100%) rename {deepspeech => paddlespeech/s2t}/transform/channel_selector.py (100%) rename {deepspeech => paddlespeech/s2t}/transform/cmvn.py (100%) rename {deepspeech => paddlespeech/s2t}/transform/functional.py (94%) rename {deepspeech => paddlespeech/s2t}/transform/perturb.py (99%) rename {deepspeech => paddlespeech/s2t}/transform/spec_augment.py (99%) rename {deepspeech => paddlespeech/s2t}/transform/spectrogram.py (100%) rename {deepspeech => paddlespeech/s2t}/transform/transform_interface.py (100%) rename {deepspeech => paddlespeech/s2t}/transform/transformation.py (78%) rename {deepspeech => paddlespeech/s2t}/transform/wpe.py (100%) rename {deepspeech => paddlespeech/s2t}/utils/__init__.py (100%) rename {deepspeech => paddlespeech/s2t}/utils/asr_utils.py (100%) rename {deepspeech => paddlespeech/s2t}/utils/bleu_score.py (100%) rename {deepspeech => paddlespeech/s2t}/utils/check_kwargs.py (100%) rename {deepspeech => paddlespeech/s2t}/utils/checkpoint.py (99%) rename {deepspeech => paddlespeech/s2t}/utils/cli_readers.py (99%) rename {deepspeech => paddlespeech/s2t}/utils/cli_utils.py (100%) rename {deepspeech => paddlespeech/s2t}/utils/cli_writers.py (98%) rename {deepspeech => paddlespeech/s2t}/utils/ctc_utils.py (98%) rename {deepspeech => paddlespeech/s2t}/utils/dynamic_import.py (89%) rename {deepspeech => paddlespeech/s2t}/utils/error_rate.py (100%) rename {deepspeech => paddlespeech/s2t}/utils/layer_tools.py (100%) rename {deepspeech => paddlespeech/s2t}/utils/log.py (100%) rename {deepspeech => paddlespeech/s2t}/utils/mp_tools.py (100%) rename {deepspeech => paddlespeech/s2t}/utils/profiler.py (99%) rename {deepspeech => paddlespeech/s2t}/utils/socket_server.py (98%) rename {deepspeech => paddlespeech/s2t}/utils/spec_augment.py (100%) rename {deepspeech => paddlespeech/s2t}/utils/tensor_utils.py (99%) rename {deepspeech => paddlespeech/s2t}/utils/text_grid.py (100%) rename {deepspeech => paddlespeech/s2t}/utils/utility.py (98%) rename {parakeet => paddlespeech/t2s}/__init__.py (100%) rename {parakeet => paddlespeech/t2s}/audio/__init__.py (100%) rename {parakeet => paddlespeech/t2s}/audio/audio.py (100%) rename {parakeet => paddlespeech/t2s}/audio/spec_normalizer.py (100%) rename {parakeet => paddlespeech/t2s}/data/__init__.py (92%) rename {parakeet => paddlespeech/t2s}/data/batch.py (100%) rename {parakeet => paddlespeech/t2s}/data/dataset.py (100%) rename {parakeet => paddlespeech/t2s}/data/get_feats.py (100%) rename {parakeet => paddlespeech/t2s}/datasets/__init__.py (100%) rename {parakeet => paddlespeech/t2s}/datasets/am_batch_fn.py (99%) rename {parakeet => paddlespeech/t2s}/datasets/common.py (100%) rename {parakeet => paddlespeech/t2s}/datasets/data_table.py (100%) rename {parakeet => paddlespeech/t2s}/datasets/ljspeech.py (100%) rename {parakeet => paddlespeech/t2s}/datasets/preprocess_utils.py (100%) rename {parakeet => paddlespeech/t2s}/datasets/vocoder_batch_fn.py (100%) rename {parakeet => paddlespeech/t2s}/exps/__init__.py (100%) rename {parakeet => paddlespeech/t2s}/exps/fastspeech2/__init__.py (100%) rename {parakeet => paddlespeech/t2s}/exps/fastspeech2/inference.py (98%) rename {parakeet => paddlespeech/t2s}/exps/fastspeech2/multi_spk_synthesize_e2e.py (93%) rename {parakeet => paddlespeech/t2s}/exps/fastspeech2/multi_spk_synthesize_e2e_en.py (94%) rename {parakeet => paddlespeech/t2s}/exps/fastspeech2/normalize.py (99%) rename {parakeet => paddlespeech/t2s}/exps/fastspeech2/preprocess.py (95%) rename {parakeet => paddlespeech/t2s}/exps/fastspeech2/synthesize.py (93%) rename {parakeet => paddlespeech/t2s}/exps/fastspeech2/synthesize_e2e.py (94%) rename {parakeet => paddlespeech/t2s}/exps/fastspeech2/synthesize_e2e_en.py (93%) rename {parakeet => paddlespeech/t2s}/exps/fastspeech2/synthesize_e2e_melgan.py (94%) rename {parakeet => paddlespeech/t2s}/exps/fastspeech2/train.py (89%) rename {parakeet => paddlespeech/t2s}/exps/gan_vocoder/README.md (100%) rename {parakeet => paddlespeech/t2s}/exps/gan_vocoder/__init__.py (100%) rename {parakeet => paddlespeech/t2s}/exps/gan_vocoder/multi_band_melgan/__init__.py (100%) rename {parakeet => paddlespeech/t2s}/exps/gan_vocoder/multi_band_melgan/synthesize.py (96%) rename {parakeet => paddlespeech/t2s}/exps/gan_vocoder/multi_band_melgan/train.py (90%) rename {parakeet => paddlespeech/t2s}/exps/gan_vocoder/normalize.py (98%) rename {parakeet => paddlespeech/t2s}/exps/gan_vocoder/parallelwave_gan/__init__.py (100%) rename {parakeet => paddlespeech/t2s}/exps/gan_vocoder/parallelwave_gan/synthesize.py (96%) rename {parakeet => paddlespeech/t2s}/exps/gan_vocoder/parallelwave_gan/synthesize_from_wav.py (93%) rename {parakeet => paddlespeech/t2s}/exps/gan_vocoder/parallelwave_gan/train.py (92%) rename {parakeet => paddlespeech/t2s}/exps/gan_vocoder/preprocess.py (98%) rename {parakeet => paddlespeech/t2s}/exps/ge2e/__init__.py (100%) rename {parakeet => paddlespeech/t2s}/exps/ge2e/audio_processor.py (100%) rename {parakeet => paddlespeech/t2s}/exps/ge2e/config.py (100%) rename {parakeet => paddlespeech/t2s}/exps/ge2e/dataset_processors.py (98%) rename {parakeet => paddlespeech/t2s}/exps/ge2e/inference.py (95%) rename {parakeet => paddlespeech/t2s}/exps/ge2e/preprocess.py (89%) rename {parakeet => paddlespeech/t2s}/exps/ge2e/random_cycle.py (100%) rename {parakeet => paddlespeech/t2s}/exps/ge2e/speaker_verification_dataset.py (98%) rename {parakeet => paddlespeech/t2s}/exps/ge2e/train.py (88%) rename {parakeet => paddlespeech/t2s}/exps/sentences.txt (100%) rename {parakeet => paddlespeech/t2s}/exps/sentences_en.txt (100%) rename {parakeet => paddlespeech/t2s}/exps/speedyspeech/__init__.py (100%) rename {parakeet => paddlespeech/t2s}/exps/speedyspeech/inference.py (98%) rename {parakeet => paddlespeech/t2s}/exps/speedyspeech/normalize.py (98%) rename {parakeet => paddlespeech/t2s}/exps/speedyspeech/preprocess.py (96%) rename {parakeet => paddlespeech/t2s}/exps/speedyspeech/synthesize.py (94%) rename {parakeet => paddlespeech/t2s}/exps/speedyspeech/synthesize_e2e.py (94%) rename {parakeet => paddlespeech/t2s}/exps/speedyspeech/train.py (91%) rename {parakeet => paddlespeech/t2s}/exps/tacotron2/__init__.py (100%) rename {parakeet => paddlespeech/t2s}/exps/tacotron2/config.py (100%) rename {parakeet => paddlespeech/t2s}/exps/tacotron2/ljspeech.py (96%) rename {parakeet => paddlespeech/t2s}/exps/tacotron2/preprocess.py (91%) rename {parakeet => paddlespeech/t2s}/exps/tacotron2/synthesize.ipynb (99%) rename {parakeet => paddlespeech/t2s}/exps/tacotron2/synthesize.py (93%) rename {parakeet => paddlespeech/t2s}/exps/tacotron2/train.py (93%) rename {parakeet => paddlespeech/t2s}/exps/transformer_tts/__init__.py (100%) rename {parakeet => paddlespeech/t2s}/exps/transformer_tts/normalize.py (98%) rename {parakeet => paddlespeech/t2s}/exps/transformer_tts/preprocess.py (98%) rename {parakeet => paddlespeech/t2s}/exps/transformer_tts/synthesize.py (92%) rename {parakeet => paddlespeech/t2s}/exps/transformer_tts/synthesize_e2e.py (93%) rename {parakeet => paddlespeech/t2s}/exps/transformer_tts/train.py (89%) rename {parakeet => paddlespeech/t2s}/exps/voice_cloning/__init__.py (100%) rename {parakeet => paddlespeech/t2s}/exps/voice_cloning/tacotron2_ge2e/__init__.py (100%) rename {parakeet => paddlespeech/t2s}/exps/voice_cloning/tacotron2_ge2e/aishell3.py (89%) rename {parakeet => paddlespeech/t2s}/exps/voice_cloning/tacotron2_ge2e/chinese_g2p.py (92%) rename {parakeet => paddlespeech/t2s}/exps/voice_cloning/tacotron2_ge2e/config.py (100%) rename {parakeet => paddlespeech/t2s}/exps/voice_cloning/tacotron2_ge2e/extract_mel.py (92%) rename {parakeet => paddlespeech/t2s}/exps/voice_cloning/tacotron2_ge2e/lexicon.txt (100%) rename {parakeet => paddlespeech/t2s}/exps/voice_cloning/tacotron2_ge2e/preprocess_transcription.py (100%) rename {parakeet => paddlespeech/t2s}/exps/voice_cloning/tacotron2_ge2e/process_wav.py (100%) rename {parakeet => paddlespeech/t2s}/exps/voice_cloning/tacotron2_ge2e/train.py (93%) rename {parakeet => paddlespeech/t2s}/exps/voice_cloning/tacotron2_ge2e/voice_cloning.py (90%) rename {parakeet => paddlespeech/t2s}/exps/waveflow/__init__.py (100%) rename {parakeet => paddlespeech/t2s}/exps/waveflow/config.py (100%) rename {parakeet => paddlespeech/t2s}/exps/waveflow/ljspeech.py (96%) rename {parakeet => paddlespeech/t2s}/exps/waveflow/preprocess.py (96%) rename {parakeet => paddlespeech/t2s}/exps/waveflow/synthesize.py (93%) rename {parakeet => paddlespeech/t2s}/exps/waveflow/train.py (89%) rename {parakeet => paddlespeech/t2s}/frontend/__init__.py (100%) rename {parakeet => paddlespeech/t2s}/frontend/arpabet.py (98%) rename {parakeet => paddlespeech/t2s}/frontend/generate_lexicon.py (100%) create mode 100644 paddlespeech/t2s/frontend/normalizer/__init__.py rename {parakeet => paddlespeech/t2s}/frontend/normalizer/abbrrviation.py (100%) rename {parakeet => paddlespeech/t2s}/frontend/normalizer/acronyms.py (100%) rename {parakeet => paddlespeech/t2s}/frontend/normalizer/normalizer.py (94%) rename {parakeet => paddlespeech/t2s}/frontend/normalizer/numbers.py (100%) rename {parakeet => paddlespeech/t2s}/frontend/normalizer/width.py (100%) rename {parakeet => paddlespeech/t2s}/frontend/phonectic.py (97%) rename {parakeet => paddlespeech/t2s}/frontend/pinyin.py (98%) rename {parakeet => paddlespeech/t2s}/frontend/punctuation.py (100%) rename {parakeet => paddlespeech/t2s}/frontend/tone_sandhi.py (100%) rename {parakeet => paddlespeech/t2s}/frontend/vocab.py (100%) rename {parakeet => paddlespeech/t2s}/frontend/zh_frontend.py (98%) rename {parakeet => paddlespeech/t2s}/frontend/zh_normalization/README.md (100%) create mode 100644 paddlespeech/t2s/frontend/zh_normalization/__init__.py rename {parakeet => paddlespeech/t2s}/frontend/zh_normalization/char_convert.py (100%) rename {parakeet => paddlespeech/t2s}/frontend/zh_normalization/chronology.py (100%) rename {parakeet => paddlespeech/t2s}/frontend/zh_normalization/constants.py (100%) rename {parakeet => paddlespeech/t2s}/frontend/zh_normalization/num.py (100%) rename {parakeet => paddlespeech/t2s}/frontend/zh_normalization/phonecode.py (100%) rename {parakeet => paddlespeech/t2s}/frontend/zh_normalization/quantifier.py (100%) rename {parakeet => paddlespeech/t2s}/frontend/zh_normalization/text_normlization.py (100%) rename {parakeet => paddlespeech/t2s}/models/__init__.py (100%) rename {parakeet => paddlespeech/t2s}/models/fastspeech2/__init__.py (100%) rename {parakeet => paddlespeech/t2s}/models/fastspeech2/fastspeech2.py (97%) rename {parakeet => paddlespeech/t2s}/models/fastspeech2/fastspeech2_updater.py (95%) rename {parakeet => paddlespeech/t2s}/models/lstm_speaker_encoder.py (100%) rename {parakeet => paddlespeech/t2s}/models/melgan/__init__.py (100%) rename {parakeet => paddlespeech/t2s}/models/melgan/melgan.py (98%) rename {parakeet => paddlespeech/t2s}/models/melgan/multi_band_melgan_updater.py (97%) rename {parakeet => paddlespeech/t2s}/models/parallel_wavegan/__init__.py (100%) rename {parakeet => paddlespeech/t2s}/models/parallel_wavegan/parallel_wavegan.py (100%) rename {parakeet => paddlespeech/t2s}/models/parallel_wavegan/parallel_wavegan_updater.py (96%) rename {parakeet => paddlespeech/t2s}/models/speedyspeech/__init__.py (100%) rename {parakeet => paddlespeech/t2s}/models/speedyspeech/speedyspeech.py (98%) rename {parakeet => paddlespeech/t2s}/models/speedyspeech/speedyspeech_updater.py (93%) rename {parakeet => paddlespeech/t2s}/models/tacotron2.py (99%) rename {parakeet => paddlespeech/t2s}/models/transformer_tts/__init__.py (100%) rename {parakeet => paddlespeech/t2s}/models/transformer_tts/transformer_tts.py (97%) rename {parakeet => paddlespeech/t2s}/models/transformer_tts/transformer_tts_updater.py (97%) rename {parakeet => paddlespeech/t2s}/models/waveflow.py (99%) rename {parakeet => paddlespeech/t2s}/modules/__init__.py (100%) rename {parakeet => paddlespeech/t2s}/modules/adversarial_loss.py (100%) rename {parakeet => paddlespeech/t2s}/modules/attention.py (100%) rename {parakeet => paddlespeech/t2s}/modules/audio.py (100%) rename {parakeet => paddlespeech/t2s}/modules/causal_conv.py (100%) rename {parakeet => paddlespeech/t2s}/modules/conv.py (100%) rename {parakeet => paddlespeech/t2s}/modules/expansion.py (100%) rename {parakeet => paddlespeech/t2s}/modules/fastspeech2_predictor/__init__.py (100%) rename {parakeet => paddlespeech/t2s}/modules/fastspeech2_predictor/duration_predictor.py (97%) rename {parakeet => paddlespeech/t2s}/modules/fastspeech2_predictor/length_regulator.py (100%) rename {parakeet => paddlespeech/t2s}/modules/fastspeech2_predictor/variance_predictor.py (96%) rename {parakeet => paddlespeech/t2s}/modules/fastspeech2_transformer/__init__.py (100%) rename {parakeet => paddlespeech/t2s}/modules/fastspeech2_transformer/attention.py (98%) rename {parakeet => paddlespeech/t2s}/modules/fastspeech2_transformer/decoder.py (93%) rename {parakeet => paddlespeech/t2s}/modules/fastspeech2_transformer/decoder_layer.py (98%) rename {parakeet => paddlespeech/t2s}/modules/fastspeech2_transformer/embedding.py (100%) rename {parakeet => paddlespeech/t2s}/modules/fastspeech2_transformer/encoder.py (92%) rename {parakeet => paddlespeech/t2s}/modules/fastspeech2_transformer/encoder_layer.py (100%) rename {parakeet => paddlespeech/t2s}/modules/fastspeech2_transformer/lightconv.py (97%) rename {parakeet => paddlespeech/t2s}/modules/fastspeech2_transformer/mask.py (100%) rename {parakeet => paddlespeech/t2s}/modules/fastspeech2_transformer/multi_layer_conv.py (100%) rename {parakeet => paddlespeech/t2s}/modules/fastspeech2_transformer/positionwise_feed_forward.py (100%) rename {parakeet => paddlespeech/t2s}/modules/fastspeech2_transformer/repeat.py (100%) rename {parakeet => paddlespeech/t2s}/modules/geometry.py (100%) rename {parakeet => paddlespeech/t2s}/modules/glu.py (100%) rename {parakeet => paddlespeech/t2s}/modules/layer_norm.py (100%) rename {parakeet => paddlespeech/t2s}/modules/losses.py (100%) rename {parakeet => paddlespeech/t2s}/modules/masked_fill.py (100%) rename {parakeet => paddlespeech/t2s}/modules/masking.py (100%) rename {parakeet => paddlespeech/t2s}/modules/nets_utils.py (100%) rename {parakeet => paddlespeech/t2s}/modules/normalizer.py (100%) rename {parakeet => paddlespeech/t2s}/modules/positional_encoding.py (100%) rename {parakeet => paddlespeech/t2s}/modules/pqmf.py (100%) rename {parakeet => paddlespeech/t2s}/modules/residual_stack.py (98%) rename {parakeet => paddlespeech/t2s}/modules/ssim.py (100%) rename {parakeet => paddlespeech/t2s}/modules/stft_loss.py (100%) rename {parakeet => paddlespeech/t2s}/modules/style_encoder.py (98%) rename {parakeet => paddlespeech/t2s}/modules/tacotron2/__init__.py (100%) rename {parakeet => paddlespeech/t2s}/modules/tacotron2/decoder.py (100%) rename {parakeet => paddlespeech/t2s}/modules/tacotron2/encoder.py (100%) rename {parakeet => paddlespeech/t2s}/modules/transformer.py (99%) rename {parakeet => paddlespeech/t2s}/training/__init__.py (100%) rename {parakeet => paddlespeech/t2s}/training/cli.py (91%) rename {parakeet => paddlespeech/t2s}/training/default_config.py (100%) rename {parakeet => paddlespeech/t2s}/training/experiment.py (99%) rename {parakeet => paddlespeech/t2s}/training/extension.py (100%) rename {parakeet => paddlespeech/t2s}/training/extensions/__init__.py (100%) rename {parakeet => paddlespeech/t2s}/training/extensions/evaluator.py (91%) rename {parakeet => paddlespeech/t2s}/training/extensions/snapshot.py (95%) rename {parakeet => paddlespeech/t2s}/training/extensions/visualizer.py (93%) rename {parakeet => paddlespeech/t2s}/training/optimizer.py (100%) rename {parakeet => paddlespeech/t2s}/training/reporter.py (100%) rename {parakeet => paddlespeech/t2s}/training/seeding.py (100%) rename {parakeet => paddlespeech/t2s}/training/trainer.py (94%) rename {parakeet => paddlespeech/t2s}/training/trigger.py (91%) rename {parakeet => paddlespeech/t2s}/training/triggers/__init__.py (100%) rename {parakeet => paddlespeech/t2s}/training/triggers/interval_trigger.py (100%) rename {parakeet => paddlespeech/t2s}/training/triggers/limit_trigger.py (100%) rename {parakeet => paddlespeech/t2s}/training/triggers/time_trigger.py (100%) rename {parakeet => paddlespeech/t2s}/training/updater.py (100%) rename {parakeet => paddlespeech/t2s}/training/updaters/__init__.py (100%) rename {parakeet => paddlespeech/t2s}/training/updaters/standard_updater.py (97%) rename {parakeet => paddlespeech/t2s}/utils/__init__.py (100%) rename {parakeet => paddlespeech/t2s}/utils/checkpoint.py (99%) rename {parakeet => paddlespeech/t2s}/utils/display.py (100%) rename {parakeet => paddlespeech/t2s}/utils/error_rate.py (100%) rename {parakeet => paddlespeech/t2s}/utils/h5_utils.py (100%) rename {parakeet => paddlespeech/t2s}/utils/internals.py (100%) rename {parakeet => paddlespeech/t2s}/utils/layer_tools.py (100%) rename {parakeet => paddlespeech/t2s}/utils/mp_tools.py (100%) rename {parakeet => paddlespeech/t2s}/utils/profile.py (100%) rename {parakeet => paddlespeech/t2s}/utils/profiler.py (100%) rename {parakeet => paddlespeech/t2s}/utils/scheduler.py (100%) rename {parakeet => paddlespeech/t2s}/utils/timeline.py (100%) rename {text_processing => paddlespeech/text}/.gitignore (100%) rename {text_processing => paddlespeech/text}/README.md (100%) rename {text_processing => paddlespeech/text}/examples/punctuation_restoration/chinese/README.md (100%) rename {text_processing => paddlespeech/text}/examples/punctuation_restoration/chinese/conf/blstm.yaml (100%) rename {text_processing => paddlespeech/text}/examples/punctuation_restoration/chinese/conf/data_conf/chinese.yaml (100%) rename {text_processing => paddlespeech/text}/examples/punctuation_restoration/chinese/conf/train_conf/bertBLSTM_zh.yaml (100%) rename {text_processing => paddlespeech/text}/examples/punctuation_restoration/chinese/conf/train_conf/bertLinear_zh.yaml (100%) rename {text_processing => paddlespeech/text}/examples/punctuation_restoration/chinese/local/avg.sh (100%) rename {text_processing => paddlespeech/text}/examples/punctuation_restoration/chinese/local/data.sh (100%) rename {text_processing => paddlespeech/text}/examples/punctuation_restoration/chinese/local/test.sh (100%) rename {text_processing => paddlespeech/text}/examples/punctuation_restoration/chinese/local/train.sh (100%) rename {text_processing => paddlespeech/text}/examples/punctuation_restoration/chinese/path.sh (100%) rename {text_processing => paddlespeech/text}/examples/punctuation_restoration/chinese/run.sh (100%) rename {text_processing => paddlespeech/text}/examples/punctuation_restoration/english/README.md (100%) rename {text_processing => paddlespeech/text}/examples/punctuation_restoration/english/conf/data_conf/english.yaml (100%) rename {text_processing => paddlespeech/text}/examples/punctuation_restoration/english/conf/train_conf/bertBLSTM_base_en.yaml (100%) rename {text_processing => paddlespeech/text}/examples/punctuation_restoration/english/conf/train_conf/bertLinear_en.yaml (100%) rename {text_processing => paddlespeech/text}/examples/punctuation_restoration/english/local/avg.sh (100%) rename {text_processing => paddlespeech/text}/examples/punctuation_restoration/english/local/data.sh (100%) rename {text_processing => paddlespeech/text}/examples/punctuation_restoration/english/local/test.sh (100%) rename {text_processing => paddlespeech/text}/examples/punctuation_restoration/english/local/train.sh (100%) rename {text_processing => paddlespeech/text}/examples/punctuation_restoration/english/path.sh (100%) rename {text_processing => paddlespeech/text}/examples/punctuation_restoration/english/run.sh (100%) rename {text_processing => paddlespeech/text}/requirements.txt (100%) rename {text_processing => paddlespeech/text}/speechtask/punctuation_restoration/bin/avg_model.py (100%) rename {text_processing => paddlespeech/text}/speechtask/punctuation_restoration/bin/pre_data.py (100%) rename {text_processing => paddlespeech/text}/speechtask/punctuation_restoration/bin/test.py (100%) rename {text_processing => paddlespeech/text}/speechtask/punctuation_restoration/bin/train.py (100%) rename {text_processing => paddlespeech/text}/speechtask/punctuation_restoration/io/__init__.py (100%) rename {text_processing => paddlespeech/text}/speechtask/punctuation_restoration/io/collator.py (100%) rename {text_processing => paddlespeech/text}/speechtask/punctuation_restoration/io/common.py (100%) rename {text_processing => paddlespeech/text}/speechtask/punctuation_restoration/io/dataset.py (100%) rename {text_processing => paddlespeech/text}/speechtask/punctuation_restoration/model/BertBLSTM.py (100%) rename {text_processing => paddlespeech/text}/speechtask/punctuation_restoration/model/BertLinear.py (100%) rename {text_processing => paddlespeech/text}/speechtask/punctuation_restoration/model/blstm.py (100%) rename {text_processing => paddlespeech/text}/speechtask/punctuation_restoration/model/lstm.py (100%) rename {text_processing => paddlespeech/text}/speechtask/punctuation_restoration/modules/__init__.py (100%) rename {text_processing => paddlespeech/text}/speechtask/punctuation_restoration/modules/activation.py (100%) rename {text_processing => paddlespeech/text}/speechtask/punctuation_restoration/modules/attention.py (100%) rename {text_processing => paddlespeech/text}/speechtask/punctuation_restoration/modules/crf.py (100%) rename {text_processing => paddlespeech/text}/speechtask/punctuation_restoration/training/__init__.py (100%) rename {text_processing => paddlespeech/text}/speechtask/punctuation_restoration/training/loss.py (100%) rename {text_processing => paddlespeech/text}/speechtask/punctuation_restoration/training/trainer.py (100%) rename {text_processing => paddlespeech/text}/speechtask/punctuation_restoration/utils/__init__.py (100%) rename {text_processing => paddlespeech/text}/speechtask/punctuation_restoration/utils/checkpoint.py (100%) rename {text_processing => paddlespeech/text}/speechtask/punctuation_restoration/utils/default_parser.py (93%) rename {text_processing => paddlespeech/text}/speechtask/punctuation_restoration/utils/layer_tools.py (100%) rename {text_processing => paddlespeech/text}/speechtask/punctuation_restoration/utils/mp_tools.py (100%) rename {text_processing => paddlespeech/text}/speechtask/punctuation_restoration/utils/punct_pre.py (100%) rename {text_processing => paddlespeech/text}/speechtask/punctuation_restoration/utils/utility.py (100%) delete mode 100644 parakeet/frontend/normalizer/__init__.py delete mode 100644 parakeet/frontend/zh_normalization/__init__.py diff --git a/.mergify.yml b/.mergify.yml index 6ec28ae8..1e306d67 100644 --- a/.mergify.yml +++ b/.mergify.yml @@ -41,13 +41,13 @@ pull_request_rules: remove: ["conflicts"] - name: "auto add label=S2T" conditions: - - files~=^deepspeech/ + - files~=^paddlespeech/s2t/ actions: label: add: ["S2T"] - name: "auto add label=T2S" conditions: - - files~=^parakeet/ + - files~=^paddlespeech/t2s/ actions: label: add: ["T2S"] @@ -59,7 +59,7 @@ pull_request_rules: add: ["Audio"] - name: "auto add label=TextProcess" conditions: - - files~=^text_processing/ + - files~=^paddlespeech/text/ actions: label: add: ["TextProcess"] diff --git a/docs/source/asr/models_introduction.md b/docs/source/asr/models_introduction.md index c99093bd..7843a349 100644 --- a/docs/source/asr/models_introduction.md +++ b/docs/source/asr/models_introduction.md @@ -61,7 +61,7 @@ For feature extraction, three methods are implemented, which are linear (FFT wit Currently, the released deepspeech2 online model use the linear feature extraction method. ``` The code for feature extraction -vi deepspeech/frontend/featurizer/audio_featurizer.py +vi paddlespeech/s2t/frontend/featurizer/audio_featurizer.py ``` ### Encoder @@ -69,7 +69,7 @@ The encoder is composed of two 2D convolution subsampling layers and a number of The code of Encoder is in: ``` -vi deepspeech/models/ds2_online/deepspeech2.py +vi paddlespeech/s2t/models/ds2_online/deepspeech2.py ``` ### Decoder @@ -78,9 +78,9 @@ To got the character possibilities of each frame, the feature representation of The code of the decoder is in: ``` # The code of constructing the decoder in model -vi deepspeech/models/ds2_online/deepspeech2.py +vi paddlespeech/s2t/models/ds2_online/deepspeech2.py # The code of CTC Decoder -vi deepspeech/modules/ctc.py +vi paddlespeech/s2t/modules/ctc.py ``` ### Training Process @@ -169,7 +169,7 @@ For data preparation and decoder, the deepspeech2 offline model is same with the The code of encoder and decoder for deepspeech2 offline model is in: ``` -vi deepspeech/models/ds2/deepspeech2.py +vi paddlespeech/s2t/models/ds2/deepspeech2.py ``` The training process and testing process of deepspeech2 offline model is very similary to deepspeech2 online model. diff --git a/docs/source/conf.py b/docs/source/conf.py index f2f75ce3..e6431c7c 100644 --- a/docs/source/conf.py +++ b/docs/source/conf.py @@ -28,8 +28,8 @@ autodoc_mock_imports = ["soundfile", "librosa"] # -- Project information ----------------------------------------------------- project = 'paddle speech' -copyright = '2021, Deepspeech-developers' -author = 'Deepspeech-developers' +copyright = '2021, paddlespeech-developers' +author = 'paddlespeech-developers' # The full version, including alpha/beta/rc tags release = '2.1' diff --git a/docs/source/index.rst b/docs/source/index.rst index 06bc2f3f..53e5d15d 100644 --- a/docs/source/index.rst +++ b/docs/source/index.rst @@ -1,7 +1,7 @@ -Welcome to paddle Deepspeech documentation ! +Welcome to paddle PaddleSpeech documentation ! ============================================== -**Deepspeech** is a Speech toolkits implemented by paddlepaddle. +**PaddleSpeech** is a Speech toolkits implemented by paddlepaddle. Contents diff --git a/docs/source/tts/advanced_usage.md b/docs/source/tts/advanced_usage.md index 297f274f..0540a1c3 100644 --- a/docs/source/tts/advanced_usage.md +++ b/docs/source/tts/advanced_usage.md @@ -67,7 +67,7 @@ There are two common ways to define a model which consists of several modules. ``` When a model is a complicated and made up of several components, each of which has a separate functionality, and can be replaced by other components with the same functionality, we prefer to define it in this way. -In the directory structure of PaddleSpeech TTS, modules with high reusability are placed in `parakeet.modules`, but models for specific tasks are placed in `parakeet.models`. When developing a new model, developers need to consider the feasibility of splitting the modules, and the degree of generality of the modules, and place them in appropriate directories. +In the directory structure of PaddleSpeech TTS, modules with high reusability are placed in `paddlespeech.t2s.modules`, but models for specific tasks are placed in `paddlespeech.t2s.models`. When developing a new model, developers need to consider the feasibility of splitting the modules, and the degree of generality of the modules, and place them in appropriate directories. ## PaddleSpeech TTS's Data Components Another critical componnet for a deep learning project is data. @@ -93,7 +93,7 @@ Then we need to select a format for saving metadata to the hard disk. There are Meanwhile, `cache` is added here, and a multi-process Manager is used to share memory between multiple processes. When `num_workers` is used, it is guaranteed that each sub process will not cache a copy. -The implementation of `DataTable` can be found in `parakeet/datasets/data_table.py`. +The implementation of `DataTable` can be found in `paddlespeech/t2s/datasets/data_table.py`. ```python class DataTable(Dataset): """Dataset to load and convert data for general purpose. @@ -179,9 +179,9 @@ We think this method is a little ugly. We prefer to return the necessary informa It takes advantage of the globality of Python's module level variables and the effect of context manager. -There is a module level variable in `parakeet/training/reporter.py` `OBSERVATIONS`，which is a `Dict` to store key-value. +There is a module level variable in `paddlespeech/t2s/training/reporter.py` `OBSERVATIONS`，which is a `Dict` to store key-value. ```python -# parakeet/training/reporter.py +# paddlespeech/t2s/training/reporter.py @contextlib.contextmanager def scope(observations): diff --git a/docs/source/tts/quick_start.md b/docs/source/tts/quick_start.md index f5d16bbf..9fd92338 100644 --- a/docs/source/tts/quick_start.md +++ b/docs/source/tts/quick_start.md @@ -102,9 +102,9 @@ import numpy as np import paddle import yaml from yacs.config import CfgNode -from parakeet.models.fastspeech2 import FastSpeech2 -from parakeet.models.fastspeech2 import FastSpeech2Inference -from parakeet.modules.normalizer import ZScore +from paddlespeech.t2s.models.fastspeech2 import FastSpeech2 +from paddlespeech.t2s.models.fastspeech2 import FastSpeech2Inference +from paddlespeech.t2s.modules.normalizer import ZScore # examples/fastspeech2/baker/frontend.py from frontend import Frontend @@ -161,9 +161,9 @@ import paddle import soundfile as sf import yaml from yacs.config import CfgNode -from parakeet.models.parallel_wavegan import PWGGenerator -from parakeet.models.parallel_wavegan import PWGInference -from parakeet.modules.normalizer import ZScore +from paddlespeech.t2s.models.parallel_wavegan import PWGGenerator +from paddlespeech.t2s.models.parallel_wavegan import PWGInference +from paddlespeech.t2s.modules.normalizer import ZScore # load the pretrained model checkpoint_dir = Path("parallel_wavegan_baker_ckpt_0.4") diff --git a/examples/aishell/s0/path.sh b/examples/aishell/s0/path.sh index a066676e..867cfb5d 100644 --- a/examples/aishell/s0/path.sh +++ b/examples/aishell/s0/path.sh @@ -11,4 +11,4 @@ export PYTHONPATH=${MAIN_ROOT}:${PYTHONPATH} export LD_LIBRARY_PATH=${LD_LIBRARY_PATH}:/usr/local/lib/ MODEL=deepspeech2 -export BIN_DIR=${MAIN_ROOT}/deepspeech/exps/${MODEL}/bin +export BIN_DIR=${MAIN_ROOT}/paddlespeech/s2t/exps/${MODEL}/bin diff --git a/examples/aishell/s1/path.sh b/examples/aishell/s1/path.sh index 0b9b0f8f..44982910 100644 --- a/examples/aishell/s1/path.sh +++ b/examples/aishell/s1/path.sh @@ -12,7 +12,7 @@ export LD_LIBRARY_PATH=${LD_LIBRARY_PATH}:/usr/local/lib/ # model exp MODEL=u2 -export BIN_DIR=${MAIN_ROOT}/deepspeech/exps/${MODEL}/bin +export BIN_DIR=${MAIN_ROOT}/paddlespeech/s2t/exps/${MODEL}/bin # srilm diff --git a/examples/aishell3/tts3/path.sh b/examples/aishell3/tts3/path.sh index 561d0163..fb7e8411 100755 --- a/examples/aishell3/tts3/path.sh +++ b/examples/aishell3/tts3/path.sh @@ -10,4 +10,4 @@ export PYTHONIOENCODING=UTF-8 export PYTHONPATH=${MAIN_ROOT}:${PYTHONPATH} MODEL=fastspeech2 -export BIN_DIR=${MAIN_ROOT}/parakeet/exps/${MODEL} +export BIN_DIR=${MAIN_ROOT}/paddlespeech/t2s/exps/${MODEL} diff --git a/examples/aishell3/vc0/README.md b/examples/aishell3/vc0/README.md index 9364cf00..9a269ed5 100644 --- a/examples/aishell3/vc0/README.md +++ b/examples/aishell3/vc0/README.md @@ -39,7 +39,7 @@ There are silence in the edge of AISHELL-3's wavs, and the audio amplitude is ve We use Montreal Force Aligner 1.0. The label in aishell3 include pinyin，so the lexicon we provided to MFA is pinyin rather than Chinese characters. And the prosody marks(`$` and `%`) need to be removed. You shoud preprocess the dataset into the format which MFA needs, the texts have the same name with wavs and have the suffix `.lab`. -We use [lexicon.txt](https://github.com/PaddlePaddle/DeepSpeech/blob/develop/parakeet/exps/voice_cloning/tacotron2_ge2e/lexicon.txt) as the lexicon. +We use [lexicon.txt](https://github.com/PaddlePaddle/DeepSpeech/blob/develop/paddlespeech/t2s/exps/voice_cloning/tacotron2_ge2e/lexicon.txt) as the lexicon. You can download the alignment results from here [alignment_aishell3.tar.gz](https://paddlespeech.bj.bcebos.com/Parakeet/alignment_aishell3.tar.gz), or train your own MFA model reference to [use_mfa example](https://github.com/PaddlePaddle/DeepSpeech/tree/develop/examples/other/use_mfa) (use MFA1.x now) of our repo. diff --git a/examples/aishell3/vc0/path.sh b/examples/aishell3/vc0/path.sh index 485d73bf..dfae49af 100755 --- a/examples/aishell3/vc0/path.sh +++ b/examples/aishell3/vc0/path.sh @@ -10,4 +10,4 @@ export PYTHONIOENCODING=UTF-8 export PYTHONPATH=${MAIN_ROOT}:${PYTHONPATH} MODEL=voice_cloning/tacotron2_ge2e -export BIN_DIR=${MAIN_ROOT}/parakeet/exps/${MODEL} +export BIN_DIR=${MAIN_ROOT}/paddlespeech/t2s/exps/${MODEL} diff --git a/examples/callcenter/s1/path.sh b/examples/callcenter/s1/path.sh index e4e278da..6d95864b 100644 --- a/examples/callcenter/s1/path.sh +++ b/examples/callcenter/s1/path.sh @@ -12,4 +12,4 @@ export LD_LIBRARY_PATH=${LD_LIBRARY_PATH}:/usr/local/lib/ MODEL=u2 -export BIN_DIR=${MAIN_ROOT}/deepspeech/exps/${MODEL}/bin +export BIN_DIR=${MAIN_ROOT}/paddlespeech/s2t/exps/${MODEL}/bin diff --git a/examples/csmsc/tts2/path.sh b/examples/csmsc/tts2/path.sh index 1a9519f3..ad0dd552 100755 --- a/examples/csmsc/tts2/path.sh +++ b/examples/csmsc/tts2/path.sh @@ -10,4 +10,4 @@ export PYTHONIOENCODING=UTF-8 export PYTHONPATH=${MAIN_ROOT}:${PYTHONPATH} MODEL=speedyspeech -export BIN_DIR=${MAIN_ROOT}/parakeet/exps/${MODEL} \ No newline at end of file +export BIN_DIR=${MAIN_ROOT}/paddlespeech/t2s/exps/${MODEL} \ No newline at end of file diff --git a/examples/csmsc/tts3/path.sh b/examples/csmsc/tts3/path.sh index 561d0163..fb7e8411 100755 --- a/examples/csmsc/tts3/path.sh +++ b/examples/csmsc/tts3/path.sh @@ -10,4 +10,4 @@ export PYTHONIOENCODING=UTF-8 export PYTHONPATH=${MAIN_ROOT}:${PYTHONPATH} MODEL=fastspeech2 -export BIN_DIR=${MAIN_ROOT}/parakeet/exps/${MODEL} +export BIN_DIR=${MAIN_ROOT}/paddlespeech/t2s/exps/${MODEL} diff --git a/examples/csmsc/voc1/path.sh b/examples/csmsc/voc1/path.sh index 28d39ae0..1e6647b8 100755 --- a/examples/csmsc/voc1/path.sh +++ b/examples/csmsc/voc1/path.sh @@ -10,4 +10,4 @@ export PYTHONIOENCODING=UTF-8 export PYTHONPATH=${MAIN_ROOT}:${PYTHONPATH} MODEL=parallelwave_gan -export BIN_DIR=${MAIN_ROOT}/parakeet/exps/gan_vocoder/${MODEL} \ No newline at end of file +export BIN_DIR=${MAIN_ROOT}/paddlespeech/t2s/exps/gan_vocoder/${MODEL} \ No newline at end of file diff --git a/examples/csmsc/voc3/path.sh b/examples/csmsc/voc3/path.sh index f6b9fe61..86c8784a 100755 --- a/examples/csmsc/voc3/path.sh +++ b/examples/csmsc/voc3/path.sh @@ -10,4 +10,4 @@ export PYTHONIOENCODING=UTF-8 export PYTHONPATH=${MAIN_ROOT}:${PYTHONPATH} MODEL=multi_band_melgan -export BIN_DIR=${MAIN_ROOT}/parakeet/exps/gan_vocoder/${MODEL} \ No newline at end of file +export BIN_DIR=${MAIN_ROOT}/paddlespeech/t2s/exps/gan_vocoder/${MODEL} \ No newline at end of file diff --git a/examples/librispeech/s0/path.sh b/examples/librispeech/s0/path.sh index 5ec461d7..0538d6f1 100644 --- a/examples/librispeech/s0/path.sh +++ b/examples/librispeech/s0/path.sh @@ -12,4 +12,4 @@ export LD_LIBRARY_PATH=${LD_LIBRARY_PATH}:/usr/local/lib/ MODEL=deepspeech2 -export BIN_DIR=${MAIN_ROOT}/deepspeech/exps/${MODEL}/bin +export BIN_DIR=${MAIN_ROOT}/paddlespeech/s2t/exps/${MODEL}/bin diff --git a/examples/librispeech/s1/path.sh b/examples/librispeech/s1/path.sh index 581a350a..e825979c 100644 --- a/examples/librispeech/s1/path.sh +++ b/examples/librispeech/s1/path.sh @@ -12,4 +12,4 @@ export LD_LIBRARY_PATH=${LD_LIBRARY_PATH}:/usr/local/lib/ MODEL=u2 -export BIN_DIR=${MAIN_ROOT}/deepspeech/exps/${MODEL}/bin +export BIN_DIR=${MAIN_ROOT}/paddlespeech/s2t/exps/${MODEL}/bin diff --git a/examples/librispeech/s2/path.sh b/examples/librispeech/s2/path.sh index 32ff28c1..ad6b6913 100644 --- a/examples/librispeech/s2/path.sh +++ b/examples/librispeech/s2/path.sh @@ -12,7 +12,7 @@ export LD_LIBRARY_PATH=${LD_LIBRARY_PATH}:/usr/local/lib/ MODEL=u2_kaldi -export BIN_DIR=${MAIN_ROOT}/deepspeech/exps/${MODEL}/bin +export BIN_DIR=${MAIN_ROOT}/paddlespeech/s2t/exps/${MODEL}/bin # srilm export LIBLBFGS=${MAIN_ROOT}/tools/liblbfgs-1.10 diff --git a/examples/ljspeech/tts0/path.sh b/examples/ljspeech/tts0/path.sh index 590e7a9c..a37cd21e 100755 --- a/examples/ljspeech/tts0/path.sh +++ b/examples/ljspeech/tts0/path.sh @@ -10,4 +10,4 @@ export PYTHONIOENCODING=UTF-8 export PYTHONPATH=${MAIN_ROOT}:${PYTHONPATH} MODEL=tacotron2 -export BIN_DIR=${MAIN_ROOT}/parakeet/exps/${MODEL} +export BIN_DIR=${MAIN_ROOT}/paddlespeech/t2s/exps/${MODEL} diff --git a/examples/ljspeech/tts1/path.sh b/examples/ljspeech/tts1/path.sh index 201261b1..32eecd85 100755 --- a/examples/ljspeech/tts1/path.sh +++ b/examples/ljspeech/tts1/path.sh @@ -10,4 +10,4 @@ export PYTHONIOENCODING=UTF-8 export PYTHONPATH=${MAIN_ROOT}:${PYTHONPATH} MODEL=transformer_tts -export BIN_DIR=${MAIN_ROOT}/parakeet/exps/${MODEL} +export BIN_DIR=${MAIN_ROOT}/paddlespeech/t2s/exps/${MODEL} diff --git a/examples/ljspeech/tts3/path.sh b/examples/ljspeech/tts3/path.sh index 561d0163..fb7e8411 100755 --- a/examples/ljspeech/tts3/path.sh +++ b/examples/ljspeech/tts3/path.sh @@ -10,4 +10,4 @@ export PYTHONIOENCODING=UTF-8 export PYTHONPATH=${MAIN_ROOT}:${PYTHONPATH} MODEL=fastspeech2 -export BIN_DIR=${MAIN_ROOT}/parakeet/exps/${MODEL} +export BIN_DIR=${MAIN_ROOT}/paddlespeech/t2s/exps/${MODEL} diff --git a/examples/ljspeech/voc0/path.sh b/examples/ljspeech/voc0/path.sh index b9fe83ec..4705c9e1 100755 --- a/examples/ljspeech/voc0/path.sh +++ b/examples/ljspeech/voc0/path.sh @@ -10,4 +10,4 @@ export PYTHONIOENCODING=UTF-8 export PYTHONPATH=${MAIN_ROOT}:${PYTHONPATH} MODEL=waveflow -export BIN_DIR=${MAIN_ROOT}/parakeet/exps/${MODEL} \ No newline at end of file +export BIN_DIR=${MAIN_ROOT}/paddlespeech/t2s/exps/${MODEL} \ No newline at end of file diff --git a/examples/ljspeech/voc1/path.sh b/examples/ljspeech/voc1/path.sh index 28d39ae0..1e6647b8 100755 --- a/examples/ljspeech/voc1/path.sh +++ b/examples/ljspeech/voc1/path.sh @@ -10,4 +10,4 @@ export PYTHONIOENCODING=UTF-8 export PYTHONPATH=${MAIN_ROOT}:${PYTHONPATH} MODEL=parallelwave_gan -export BIN_DIR=${MAIN_ROOT}/parakeet/exps/gan_vocoder/${MODEL} \ No newline at end of file +export BIN_DIR=${MAIN_ROOT}/paddlespeech/t2s/exps/gan_vocoder/${MODEL} \ No newline at end of file diff --git a/examples/other/1xt2x/src_deepspeech2x/__init__.py b/examples/other/1xt2x/src_deepspeech2x/__init__.py index d85a3dde..62f3eee3 100644 --- a/examples/other/1xt2x/src_deepspeech2x/__init__.py +++ b/examples/other/1xt2x/src_deepspeech2x/__init__.py @@ -21,7 +21,7 @@ from paddle import nn from paddle.fluid import core from paddle.nn import functional as F -from deepspeech.utils.log import Log +from paddlespeech.s2t.utils.log import Log #TODO(Hui Zhang): remove fluid import logger = Log(__name__).getlog() diff --git a/examples/other/1xt2x/src_deepspeech2x/bin/test.py b/examples/other/1xt2x/src_deepspeech2x/bin/test.py index be3f9822..b4f9cdf9 100644 --- a/examples/other/1xt2x/src_deepspeech2x/bin/test.py +++ b/examples/other/1xt2x/src_deepspeech2x/bin/test.py @@ -14,9 +14,9 @@ """Evaluation for DeepSpeech2 model.""" from src_deepspeech2x.test_model import DeepSpeech2Tester as Tester -from deepspeech.exps.deepspeech2.config import get_cfg_defaults -from deepspeech.training.cli import default_argument_parser -from deepspeech.utils.utility import print_arguments +from paddlespeech.s2t.exps.deepspeech2.config import get_cfg_defaults +from paddlespeech.s2t.training.cli import default_argument_parser +from paddlespeech.s2t.utils.utility import print_arguments def main_sp(config, args): diff --git a/examples/other/1xt2x/src_deepspeech2x/models/ds2/deepspeech2.py b/examples/other/1xt2x/src_deepspeech2x/models/ds2/deepspeech2.py index ce123743..ad83a41d 100644 --- a/examples/other/1xt2x/src_deepspeech2x/models/ds2/deepspeech2.py +++ b/examples/other/1xt2x/src_deepspeech2x/models/ds2/deepspeech2.py @@ -19,11 +19,11 @@ from paddle import nn from src_deepspeech2x.models.ds2.rnn import RNNStack from yacs.config import CfgNode -from deepspeech.models.ds2.conv import ConvStack -from deepspeech.modules.ctc import CTCDecoder -from deepspeech.utils import layer_tools -from deepspeech.utils.checkpoint import Checkpoint -from deepspeech.utils.log import Log +from paddlespeech.s2t.models.ds2.conv import ConvStack +from paddlespeech.s2t.modules.ctc import CTCDecoder +from paddlespeech.s2t.utils import layer_tools +from paddlespeech.s2t.utils.checkpoint import Checkpoint +from paddlespeech.s2t.utils.log import Log logger = Log(__name__).getlog() __all__ = ['DeepSpeech2Model', 'DeepSpeech2InferModel'] diff --git a/examples/other/1xt2x/src_deepspeech2x/models/ds2/rnn.py b/examples/other/1xt2x/src_deepspeech2x/models/ds2/rnn.py index e45db7c0..383a0746 100644 --- a/examples/other/1xt2x/src_deepspeech2x/models/ds2/rnn.py +++ b/examples/other/1xt2x/src_deepspeech2x/models/ds2/rnn.py @@ -18,9 +18,9 @@ from paddle import nn from paddle.nn import functional as F from paddle.nn import initializer as I -from deepspeech.modules.activation import brelu -from deepspeech.modules.mask import make_non_pad_mask -from deepspeech.utils.log import Log +from paddlespeech.s2t.modules.activation import brelu +from paddlespeech.s2t.modules.mask import make_non_pad_mask +from paddlespeech.s2t.utils.log import Log logger = Log(__name__).getlog() __all__ = ['RNNStack'] diff --git a/examples/other/1xt2x/src_deepspeech2x/test_model.py b/examples/other/1xt2x/src_deepspeech2x/test_model.py index 203a3bac..a9afc631 100644 --- a/examples/other/1xt2x/src_deepspeech2x/test_model.py +++ b/examples/other/1xt2x/src_deepspeech2x/test_model.py @@ -26,19 +26,19 @@ from src_deepspeech2x.models.ds2 import DeepSpeech2InferModel from src_deepspeech2x.models.ds2 import DeepSpeech2Model from yacs.config import CfgNode -from deepspeech.frontend.featurizer.text_featurizer import TextFeaturizer -from deepspeech.io.collator import SpeechCollator -from deepspeech.io.dataset import ManifestDataset -from deepspeech.io.sampler import SortagradBatchSampler -from deepspeech.io.sampler import SortagradDistributedBatchSampler -from deepspeech.models.ds2_online import DeepSpeech2InferModelOnline -from deepspeech.models.ds2_online import DeepSpeech2ModelOnline -from deepspeech.training.gradclip import ClipGradByGlobalNormWithLog -from deepspeech.training.trainer import Trainer -from deepspeech.utils import error_rate -from deepspeech.utils import layer_tools -from deepspeech.utils import mp_tools -from deepspeech.utils.log import Log +from paddlespeech.s2t.frontend.featurizer.text_featurizer import TextFeaturizer +from paddlespeech.s2t.io.collator import SpeechCollator +from paddlespeech.s2t.io.dataset import ManifestDataset +from paddlespeech.s2t.io.sampler import SortagradBatchSampler +from paddlespeech.s2t.io.sampler import SortagradDistributedBatchSampler +from paddlespeech.s2t.models.ds2_online import DeepSpeech2InferModelOnline +from paddlespeech.s2t.models.ds2_online import DeepSpeech2ModelOnline +from paddlespeech.s2t.training.gradclip import ClipGradByGlobalNormWithLog +from paddlespeech.s2t.training.trainer import Trainer +from paddlespeech.s2t.utils import error_rate +from paddlespeech.s2t.utils import layer_tools +from paddlespeech.s2t.utils import mp_tools +from paddlespeech.s2t.utils.log import Log logger = Log(__name__).getlog() diff --git a/examples/other/ge2e/path.sh b/examples/other/ge2e/path.sh index 4333199c..b4f77985 100755 --- a/examples/other/ge2e/path.sh +++ b/examples/other/ge2e/path.sh @@ -10,4 +10,4 @@ export PYTHONIOENCODING=UTF-8 export PYTHONPATH=${MAIN_ROOT}:${PYTHONPATH} MODEL=ge2e -export BIN_DIR=${MAIN_ROOT}/parakeet/exps/${MODEL} +export BIN_DIR=${MAIN_ROOT}/paddlespeech/t2s/exps/${MODEL} diff --git a/examples/other/text_frontend/test_g2p.py b/examples/other/text_frontend/test_g2p.py index 15005a00..2f092086 100644 --- a/examples/other/text_frontend/test_g2p.py +++ b/examples/other/text_frontend/test_g2p.py @@ -15,8 +15,8 @@ import argparse import re from pathlib import Path -from parakeet.frontend.zh_frontend import Frontend as zhFrontend -from parakeet.utils.error_rate import word_errors +from paddlespeech.t2s.frontend.zh_frontend import Frontend as zhFrontend +from paddlespeech.t2s.utils.error_rate import word_errors SILENCE_TOKENS = {"sp", "sil", "sp1", "spl"} diff --git a/examples/other/text_frontend/test_textnorm.py b/examples/other/text_frontend/test_textnorm.py index 22f90f87..503f959e 100644 --- a/examples/other/text_frontend/test_textnorm.py +++ b/examples/other/text_frontend/test_textnorm.py @@ -15,8 +15,8 @@ import argparse import re from pathlib import Path -from parakeet.frontend.zh_normalization.text_normlization import TextNormalizer -from parakeet.utils.error_rate import char_errors +from paddlespeech.t2s.frontend.zh_normalization.text_normlization import TextNormalizer +from paddlespeech.t2s.utils.error_rate import char_errors # delete english characters diff --git a/examples/ted_en_zh/t0/path.sh b/examples/ted_en_zh/t0/path.sh index f6801e43..fd537917 100644 --- a/examples/ted_en_zh/t0/path.sh +++ b/examples/ted_en_zh/t0/path.sh @@ -12,4 +12,4 @@ export LD_LIBRARY_PATH=${LD_LIBRARY_PATH}:/usr/local/lib/ MODEL=u2_st -export BIN_DIR=${MAIN_ROOT}/deepspeech/exps/${MODEL}/bin +export BIN_DIR=${MAIN_ROOT}/paddlespeech/s2t/exps/${MODEL}/bin diff --git a/examples/timit/s1/path.sh b/examples/timit/s1/path.sh index e4e278da..6d95864b 100644 --- a/examples/timit/s1/path.sh +++ b/examples/timit/s1/path.sh @@ -12,4 +12,4 @@ export LD_LIBRARY_PATH=${LD_LIBRARY_PATH}:/usr/local/lib/ MODEL=u2 -export BIN_DIR=${MAIN_ROOT}/deepspeech/exps/${MODEL}/bin +export BIN_DIR=${MAIN_ROOT}/paddlespeech/s2t/exps/${MODEL}/bin diff --git a/examples/tiny/s0/path.sh b/examples/tiny/s0/path.sh index 8a9345f2..f43a623e 100644 --- a/examples/tiny/s0/path.sh +++ b/examples/tiny/s0/path.sh @@ -11,4 +11,4 @@ export LD_LIBRARY_PATH=${LD_LIBRARY_PATH}:/usr/local/lib/ MODEL=deepspeech2 -export BIN_DIR=${MAIN_ROOT}/deepspeech/exps/${MODEL}/bin +export BIN_DIR=${MAIN_ROOT}/paddlespeech/s2t/exps/${MODEL}/bin diff --git a/examples/tiny/s1/path.sh b/examples/tiny/s1/path.sh index 29841bc1..900156a6 100644 --- a/examples/tiny/s1/path.sh +++ b/examples/tiny/s1/path.sh @@ -11,4 +11,4 @@ export LD_LIBRARY_PATH=${LD_LIBRARY_PATH}:/usr/local/lib/ MODEL=u2 -export BIN_DIR=${MAIN_ROOT}/deepspeech/exps/${MODEL}/bin +export BIN_DIR=${MAIN_ROOT}/paddlespeech/s2t/exps/${MODEL}/bin diff --git a/examples/vctk/tts3/path.sh b/examples/vctk/tts3/path.sh index 561d0163..fb7e8411 100755 --- a/examples/vctk/tts3/path.sh +++ b/examples/vctk/tts3/path.sh @@ -10,4 +10,4 @@ export PYTHONIOENCODING=UTF-8 export PYTHONPATH=${MAIN_ROOT}:${PYTHONPATH} MODEL=fastspeech2 -export BIN_DIR=${MAIN_ROOT}/parakeet/exps/${MODEL} +export BIN_DIR=${MAIN_ROOT}/paddlespeech/t2s/exps/${MODEL} diff --git a/examples/vctk/voc1/path.sh b/examples/vctk/voc1/path.sh index 28d39ae0..1e6647b8 100755 --- a/examples/vctk/voc1/path.sh +++ b/examples/vctk/voc1/path.sh @@ -10,4 +10,4 @@ export PYTHONIOENCODING=UTF-8 export PYTHONPATH=${MAIN_ROOT}:${PYTHONPATH} MODEL=parallelwave_gan -export BIN_DIR=${MAIN_ROOT}/parakeet/exps/gan_vocoder/${MODEL} \ No newline at end of file +export BIN_DIR=${MAIN_ROOT}/paddlespeech/t2s/exps/gan_vocoder/${MODEL} \ No newline at end of file diff --git a/hub/setup_hub.sh b/hub/setup_hub.sh index f2d43ad1..25c89f2b 100644 --- a/hub/setup_hub.sh +++ b/hub/setup_hub.sh @@ -52,7 +52,7 @@ fi # install decoders python3 -c "import pkg_resources; pkg_resources.require(\"swig_decoders==1.1\")" if [ $? != 0 ]; then - cd deepspeech/decoders/swig > /dev/null + cd paddlespeech/s2t/decoders/swig > /dev/null sh setup.sh cd - > /dev/null fi diff --git a/paddlespeech/__init__.py b/paddlespeech/__init__.py new file mode 100644 index 00000000..e69de29b diff --git a/deepspeech/__init__.py b/paddlespeech/s2t/__init__.py similarity index 99% rename from deepspeech/__init__.py rename to paddlespeech/s2t/__init__.py index 5762e635..855ceef9 100644 --- a/deepspeech/__init__.py +++ b/paddlespeech/s2t/__init__.py @@ -21,7 +21,7 @@ from paddle import nn from paddle.fluid import core from paddle.nn import functional as F -from deepspeech.utils.log import Log +from paddlespeech.s2t.utils.log import Log #TODO(Hui Zhang): remove fluid import logger = Log(__name__).getlog() diff --git a/deepspeech/decoders/README.md b/paddlespeech/s2t/decoders/README.md similarity index 100% rename from deepspeech/decoders/README.md rename to paddlespeech/s2t/decoders/README.md diff --git a/deepspeech/decoders/__init__.py b/paddlespeech/s2t/decoders/__init__.py similarity index 100% rename from deepspeech/decoders/__init__.py rename to paddlespeech/s2t/decoders/__init__.py diff --git a/deepspeech/decoders/beam_search/__init__.py b/paddlespeech/s2t/decoders/beam_search/__init__.py similarity index 100% rename from deepspeech/decoders/beam_search/__init__.py rename to paddlespeech/s2t/decoders/beam_search/__init__.py diff --git a/deepspeech/decoders/beam_search/batch_beam_search.py b/paddlespeech/s2t/decoders/beam_search/batch_beam_search.py similarity index 100% rename from deepspeech/decoders/beam_search/batch_beam_search.py rename to paddlespeech/s2t/decoders/beam_search/batch_beam_search.py diff --git a/deepspeech/decoders/beam_search/beam_search.py b/paddlespeech/s2t/decoders/beam_search/beam_search.py similarity index 99% rename from deepspeech/decoders/beam_search/beam_search.py rename to paddlespeech/s2t/decoders/beam_search/beam_search.py index 8fd8f9b8..6c4f069d 100644 --- a/deepspeech/decoders/beam_search/beam_search.py +++ b/paddlespeech/s2t/decoders/beam_search/beam_search.py @@ -25,7 +25,7 @@ import paddle from ..scorers.scorer_interface import PartialScorerInterface from ..scorers.scorer_interface import ScorerInterface from ..utils import end_detect -from deepspeech.utils.log import Log +from paddlespeech.s2t.utils.log import Log logger = Log(__name__).getlog() diff --git a/deepspeech/decoders/ctcdecoder/__init__.py b/paddlespeech/s2t/decoders/ctcdecoder/__init__.py similarity index 100% rename from deepspeech/decoders/ctcdecoder/__init__.py rename to paddlespeech/s2t/decoders/ctcdecoder/__init__.py diff --git a/deepspeech/decoders/ctcdecoder/decoders_deprecated.py b/paddlespeech/s2t/decoders/ctcdecoder/decoders_deprecated.py similarity index 100% rename from deepspeech/decoders/ctcdecoder/decoders_deprecated.py rename to paddlespeech/s2t/decoders/ctcdecoder/decoders_deprecated.py diff --git a/deepspeech/decoders/ctcdecoder/scorer_deprecated.py b/paddlespeech/s2t/decoders/ctcdecoder/scorer_deprecated.py similarity index 100% rename from deepspeech/decoders/ctcdecoder/scorer_deprecated.py rename to paddlespeech/s2t/decoders/ctcdecoder/scorer_deprecated.py diff --git a/deepspeech/decoders/ctcdecoder/swig/.gitignore b/paddlespeech/s2t/decoders/ctcdecoder/swig/.gitignore similarity index 100% rename from deepspeech/decoders/ctcdecoder/swig/.gitignore rename to paddlespeech/s2t/decoders/ctcdecoder/swig/.gitignore diff --git a/deepspeech/decoders/ctcdecoder/swig/__init__.py b/paddlespeech/s2t/decoders/ctcdecoder/swig/__init__.py similarity index 100% rename from deepspeech/decoders/ctcdecoder/swig/__init__.py rename to paddlespeech/s2t/decoders/ctcdecoder/swig/__init__.py diff --git a/deepspeech/decoders/ctcdecoder/swig/ctc_beam_search_decoder.cpp b/paddlespeech/s2t/decoders/ctcdecoder/swig/ctc_beam_search_decoder.cpp similarity index 100% rename from deepspeech/decoders/ctcdecoder/swig/ctc_beam_search_decoder.cpp rename to paddlespeech/s2t/decoders/ctcdecoder/swig/ctc_beam_search_decoder.cpp diff --git a/deepspeech/decoders/ctcdecoder/swig/ctc_beam_search_decoder.h b/paddlespeech/s2t/decoders/ctcdecoder/swig/ctc_beam_search_decoder.h similarity index 100% rename from deepspeech/decoders/ctcdecoder/swig/ctc_beam_search_decoder.h rename to paddlespeech/s2t/decoders/ctcdecoder/swig/ctc_beam_search_decoder.h diff --git a/deepspeech/decoders/ctcdecoder/swig/ctc_greedy_decoder.cpp b/paddlespeech/s2t/decoders/ctcdecoder/swig/ctc_greedy_decoder.cpp similarity index 100% rename from deepspeech/decoders/ctcdecoder/swig/ctc_greedy_decoder.cpp rename to paddlespeech/s2t/decoders/ctcdecoder/swig/ctc_greedy_decoder.cpp diff --git a/deepspeech/decoders/ctcdecoder/swig/ctc_greedy_decoder.h b/paddlespeech/s2t/decoders/ctcdecoder/swig/ctc_greedy_decoder.h similarity index 100% rename from deepspeech/decoders/ctcdecoder/swig/ctc_greedy_decoder.h rename to paddlespeech/s2t/decoders/ctcdecoder/swig/ctc_greedy_decoder.h diff --git a/deepspeech/decoders/ctcdecoder/swig/decoder_utils.cpp b/paddlespeech/s2t/decoders/ctcdecoder/swig/decoder_utils.cpp similarity index 100% rename from deepspeech/decoders/ctcdecoder/swig/decoder_utils.cpp rename to paddlespeech/s2t/decoders/ctcdecoder/swig/decoder_utils.cpp diff --git a/deepspeech/decoders/ctcdecoder/swig/decoder_utils.h b/paddlespeech/s2t/decoders/ctcdecoder/swig/decoder_utils.h similarity index 100% rename from deepspeech/decoders/ctcdecoder/swig/decoder_utils.h rename to paddlespeech/s2t/decoders/ctcdecoder/swig/decoder_utils.h diff --git a/deepspeech/decoders/ctcdecoder/swig/decoders.i b/paddlespeech/s2t/decoders/ctcdecoder/swig/decoders.i similarity index 100% rename from deepspeech/decoders/ctcdecoder/swig/decoders.i rename to paddlespeech/s2t/decoders/ctcdecoder/swig/decoders.i diff --git a/deepspeech/decoders/ctcdecoder/swig/path_trie.cpp b/paddlespeech/s2t/decoders/ctcdecoder/swig/path_trie.cpp similarity index 100% rename from deepspeech/decoders/ctcdecoder/swig/path_trie.cpp rename to paddlespeech/s2t/decoders/ctcdecoder/swig/path_trie.cpp diff --git a/deepspeech/decoders/ctcdecoder/swig/path_trie.h b/paddlespeech/s2t/decoders/ctcdecoder/swig/path_trie.h similarity index 100% rename from deepspeech/decoders/ctcdecoder/swig/path_trie.h rename to paddlespeech/s2t/decoders/ctcdecoder/swig/path_trie.h diff --git a/deepspeech/decoders/ctcdecoder/swig/scorer.cpp b/paddlespeech/s2t/decoders/ctcdecoder/swig/scorer.cpp similarity index 100% rename from deepspeech/decoders/ctcdecoder/swig/scorer.cpp rename to paddlespeech/s2t/decoders/ctcdecoder/swig/scorer.cpp diff --git a/deepspeech/decoders/ctcdecoder/swig/scorer.h b/paddlespeech/s2t/decoders/ctcdecoder/swig/scorer.h similarity index 100% rename from deepspeech/decoders/ctcdecoder/swig/scorer.h rename to paddlespeech/s2t/decoders/ctcdecoder/swig/scorer.h diff --git a/deepspeech/decoders/ctcdecoder/swig/setup.py b/paddlespeech/s2t/decoders/ctcdecoder/swig/setup.py similarity index 100% rename from deepspeech/decoders/ctcdecoder/swig/setup.py rename to paddlespeech/s2t/decoders/ctcdecoder/swig/setup.py diff --git a/deepspeech/decoders/ctcdecoder/swig/setup.sh b/paddlespeech/s2t/decoders/ctcdecoder/swig/setup.sh similarity index 100% rename from deepspeech/decoders/ctcdecoder/swig/setup.sh rename to paddlespeech/s2t/decoders/ctcdecoder/swig/setup.sh diff --git a/deepspeech/decoders/ctcdecoder/swig_wrapper.py b/paddlespeech/s2t/decoders/ctcdecoder/swig_wrapper.py similarity index 100% rename from deepspeech/decoders/ctcdecoder/swig_wrapper.py rename to paddlespeech/s2t/decoders/ctcdecoder/swig_wrapper.py diff --git a/deepspeech/decoders/ctcdecoder/tests/test_decoders.py b/paddlespeech/s2t/decoders/ctcdecoder/tests/test_decoders.py similarity index 98% rename from deepspeech/decoders/ctcdecoder/tests/test_decoders.py rename to paddlespeech/s2t/decoders/ctcdecoder/tests/test_decoders.py index 6e6265ef..a284890d 100644 --- a/deepspeech/decoders/ctcdecoder/tests/test_decoders.py +++ b/paddlespeech/s2t/decoders/ctcdecoder/tests/test_decoders.py @@ -14,7 +14,7 @@ """Test decoders.""" import unittest -from deepspeech.decoders import decoders_deprecated as decoder +from paddlespeech.s2t.decoders import decoders_deprecated as decoder class TestDecoders(unittest.TestCase): diff --git a/deepspeech/decoders/recog.py b/paddlespeech/s2t/decoders/recog.py similarity index 95% rename from deepspeech/decoders/recog.py rename to paddlespeech/s2t/decoders/recog.py index bc48e692..d0710f9c 100644 --- a/deepspeech/decoders/recog.py +++ b/paddlespeech/s2t/decoders/recog.py @@ -21,11 +21,11 @@ from .beam_search import BeamSearch from .scorers.length_bonus import LengthBonus from .scorers.scorer_interface import BatchScorerInterface from .utils import add_results_to_json -from deepspeech.exps import dynamic_import_tester -from deepspeech.io.reader import LoadInputsAndTargets -from deepspeech.models.asr_interface import ASRInterface -from deepspeech.models.lm_interface import dynamic_import_lm -from deepspeech.utils.log import Log +from paddlespeech.s2t.exps import dynamic_import_tester +from paddlespeech.s2t.io.reader import LoadInputsAndTargets +from paddlespeech.s2t.models.asr_interface import ASRInterface +from paddlespeech.s2t.models.lm_interface import dynamic_import_lm +from paddlespeech.s2t.utils.log import Log logger = Log(__name__).getlog() diff --git a/deepspeech/decoders/recog_bin.py b/paddlespeech/s2t/decoders/recog_bin.py similarity index 99% rename from deepspeech/decoders/recog_bin.py rename to paddlespeech/s2t/decoders/recog_bin.py index 7c866648..e4d81f40 100644 --- a/deepspeech/decoders/recog_bin.py +++ b/paddlespeech/s2t/decoders/recog_bin.py @@ -357,7 +357,7 @@ def main(args): if args.num_encs == 1: # Experimental API that supports custom LMs if args.api == "v2": - from deepspeech.decoders.recog import recog_v2 + from paddlespeech.s2t.decoders.recog import recog_v2 recog_v2(args) else: raise ValueError("Only support --api v2") diff --git a/deepspeech/decoders/scorers/__init__.py b/paddlespeech/s2t/decoders/scorers/__init__.py similarity index 100% rename from deepspeech/decoders/scorers/__init__.py rename to paddlespeech/s2t/decoders/scorers/__init__.py diff --git a/deepspeech/decoders/scorers/ctc.py b/paddlespeech/s2t/decoders/scorers/ctc.py similarity index 98% rename from deepspeech/decoders/scorers/ctc.py rename to paddlespeech/s2t/decoders/scorers/ctc.py index 4871d6e1..74a5f578 100644 --- a/deepspeech/decoders/scorers/ctc.py +++ b/paddlespeech/s2t/decoders/scorers/ctc.py @@ -28,7 +28,7 @@ class CTCPrefixScorer(BatchPartialScorerInterface): Args: ctc (paddle.nn.Layer): The CTC implementation. - For example, :class:`deepspeech.modules.ctc.CTC` + For example, :class:`paddlespeech.s2t.modules.ctc.CTC` eos (int): The end-of-sequence id. """ diff --git a/deepspeech/decoders/scorers/ctc_prefix_score.py b/paddlespeech/s2t/decoders/scorers/ctc_prefix_score.py similarity index 100% rename from deepspeech/decoders/scorers/ctc_prefix_score.py rename to paddlespeech/s2t/decoders/scorers/ctc_prefix_score.py diff --git a/deepspeech/decoders/scorers/length_bonus.py b/paddlespeech/s2t/decoders/scorers/length_bonus.py similarity index 100% rename from deepspeech/decoders/scorers/length_bonus.py rename to paddlespeech/s2t/decoders/scorers/length_bonus.py diff --git a/deepspeech/decoders/scorers/ngram.py b/paddlespeech/s2t/decoders/scorers/ngram.py similarity index 100% rename from deepspeech/decoders/scorers/ngram.py rename to paddlespeech/s2t/decoders/scorers/ngram.py diff --git a/deepspeech/decoders/scorers/scorer_interface.py b/paddlespeech/s2t/decoders/scorers/scorer_interface.py similarity index 100% rename from deepspeech/decoders/scorers/scorer_interface.py rename to paddlespeech/s2t/decoders/scorers/scorer_interface.py diff --git a/deepspeech/decoders/utils.py b/paddlespeech/s2t/decoders/utils.py similarity index 98% rename from deepspeech/decoders/utils.py rename to paddlespeech/s2t/decoders/utils.py index 3ed9c5da..7dff14c1 100644 --- a/deepspeech/decoders/utils.py +++ b/paddlespeech/s2t/decoders/utils.py @@ -13,7 +13,7 @@ # limitations under the License. import numpy as np -from deepspeech.utils.log import Log +from paddlespeech.s2t.utils.log import Log logger = Log(__name__).getlog() __all__ = ["end_detect", "parse_hypothesis", "add_results_to_json"] diff --git a/deepspeech/exps/__init__.py b/paddlespeech/s2t/exps/__init__.py similarity index 69% rename from deepspeech/exps/__init__.py rename to paddlespeech/s2t/exps/__init__.py index 29953014..b4d03066 100644 --- a/deepspeech/exps/__init__.py +++ b/paddlespeech/s2t/exps/__init__.py @@ -11,14 +11,14 @@ # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. -from deepspeech.training.trainer import Trainer -from deepspeech.utils.dynamic_import import dynamic_import +from paddlespeech.s2t.training.trainer import Trainer +from paddlespeech.s2t.utils.dynamic_import import dynamic_import model_trainer_alias = { - "ds2": "deepspeech.exp.deepspeech2.model:DeepSpeech2Trainer", - "u2": "deepspeech.exps.u2.model:U2Trainer", - "u2_kaldi": "deepspeech.exps.u2_kaldi.model:U2Trainer", - "u2_st": "deepspeech.exps.u2_st.model:U2STTrainer", + "ds2": "paddlespeech.s2t.exp.deepspeech2.model:DeepSpeech2Trainer", + "u2": "paddlespeech.s2t.exps.u2.model:U2Trainer", + "u2_kaldi": "paddlespeech.s2t.exps.u2_kaldi.model:U2Trainer", + "u2_st": "paddlespeech.s2t.exps.u2_st.model:U2STTrainer", } @@ -39,10 +39,10 @@ def dynamic_import_trainer(module): model_tester_alias = { - "ds2": "deepspeech.exp.deepspeech2.model:DeepSpeech2Tester", - "u2": "deepspeech.exps.u2.model:U2Tester", - "u2_kaldi": "deepspeech.exps.u2_kaldi.model:U2Tester", - "u2_st": "deepspeech.exps.u2_st.model:U2STTester", + "ds2": "paddlespeech.s2t.exp.deepspeech2.model:DeepSpeech2Tester", + "u2": "paddlespeech.s2t.exps.u2.model:U2Tester", + "u2_kaldi": "paddlespeech.s2t.exps.u2_kaldi.model:U2Tester", + "u2_st": "paddlespeech.s2t.exps.u2_st.model:U2STTester", } diff --git a/deepspeech/exps/deepspeech2/__init__.py b/paddlespeech/s2t/exps/deepspeech2/__init__.py similarity index 100% rename from deepspeech/exps/deepspeech2/__init__.py rename to paddlespeech/s2t/exps/deepspeech2/__init__.py diff --git a/deepspeech/exps/deepspeech2/bin/deploy/client.py b/paddlespeech/s2t/exps/deepspeech2/bin/deploy/client.py similarity index 97% rename from deepspeech/exps/deepspeech2/bin/deploy/client.py rename to paddlespeech/s2t/exps/deepspeech2/bin/deploy/client.py index 0a83f8a5..f7ed8429 100644 --- a/deepspeech/exps/deepspeech2/bin/deploy/client.py +++ b/paddlespeech/s2t/exps/deepspeech2/bin/deploy/client.py @@ -18,7 +18,7 @@ import sys import keyboard import pyaudio -from deepspeech.utils.socket_server import socket_send +from paddlespeech.s2t.utils.socket_server import socket_send parser = argparse.ArgumentParser(description=__doc__) parser.add_argument( diff --git a/deepspeech/exps/deepspeech2/bin/deploy/record.py b/paddlespeech/s2t/exps/deepspeech2/bin/deploy/record.py similarity index 100% rename from deepspeech/exps/deepspeech2/bin/deploy/record.py rename to paddlespeech/s2t/exps/deepspeech2/bin/deploy/record.py diff --git a/deepspeech/exps/deepspeech2/bin/deploy/runtime.py b/paddlespeech/s2t/exps/deepspeech2/bin/deploy/runtime.py similarity index 90% rename from deepspeech/exps/deepspeech2/bin/deploy/runtime.py rename to paddlespeech/s2t/exps/deepspeech2/bin/deploy/runtime.py index 21ffa6bf..7ccb3a6c 100644 --- a/deepspeech/exps/deepspeech2/bin/deploy/runtime.py +++ b/paddlespeech/s2t/exps/deepspeech2/bin/deploy/runtime.py @@ -20,16 +20,16 @@ from paddle.inference import Config from paddle.inference import create_predictor from paddle.io import DataLoader -from deepspeech.exps.deepspeech2.config import get_cfg_defaults -from deepspeech.io.collator import SpeechCollator -from deepspeech.io.dataset import ManifestDataset -from deepspeech.models.ds2 import DeepSpeech2Model -from deepspeech.training.cli import default_argument_parser -from deepspeech.utils.socket_server import AsrRequestHandler -from deepspeech.utils.socket_server import AsrTCPServer -from deepspeech.utils.socket_server import warm_up_test -from deepspeech.utils.utility import add_arguments -from deepspeech.utils.utility import print_arguments +from paddlespeech.s2t.exps.deepspeech2.config import get_cfg_defaults +from paddlespeech.s2t.io.collator import SpeechCollator +from paddlespeech.s2t.io.dataset import ManifestDataset +from paddlespeech.s2t.models.ds2 import DeepSpeech2Model +from paddlespeech.s2t.training.cli import default_argument_parser +from paddlespeech.s2t.utils.socket_server import AsrRequestHandler +from paddlespeech.s2t.utils.socket_server import AsrTCPServer +from paddlespeech.s2t.utils.socket_server import warm_up_test +from paddlespeech.s2t.utils.utility import add_arguments +from paddlespeech.s2t.utils.utility import print_arguments def init_predictor(args): diff --git a/deepspeech/exps/deepspeech2/bin/deploy/send.py b/paddlespeech/s2t/exps/deepspeech2/bin/deploy/send.py similarity index 96% rename from deepspeech/exps/deepspeech2/bin/deploy/send.py rename to paddlespeech/s2t/exps/deepspeech2/bin/deploy/send.py index ce6c7c84..596e7010 100644 --- a/deepspeech/exps/deepspeech2/bin/deploy/send.py +++ b/paddlespeech/s2t/exps/deepspeech2/bin/deploy/send.py @@ -15,7 +15,7 @@ import argparse import wave -from deepspeech.utils.socket_server import socket_send +from paddlespeech.s2t.utils.socket_server import socket_send parser = argparse.ArgumentParser(description=__doc__) parser.add_argument( diff --git a/deepspeech/exps/deepspeech2/bin/deploy/server.py b/paddlespeech/s2t/exps/deepspeech2/bin/deploy/server.py similarity index 87% rename from deepspeech/exps/deepspeech2/bin/deploy/server.py rename to paddlespeech/s2t/exps/deepspeech2/bin/deploy/server.py index 583e9095..5c6eee3f 100644 --- a/deepspeech/exps/deepspeech2/bin/deploy/server.py +++ b/paddlespeech/s2t/exps/deepspeech2/bin/deploy/server.py @@ -18,16 +18,16 @@ import numpy as np import paddle from paddle.io import DataLoader -from deepspeech.exps.deepspeech2.config import get_cfg_defaults -from deepspeech.io.collator import SpeechCollator -from deepspeech.io.dataset import ManifestDataset -from deepspeech.models.ds2 import DeepSpeech2Model -from deepspeech.training.cli import default_argument_parser -from deepspeech.utils.socket_server import AsrRequestHandler -from deepspeech.utils.socket_server import AsrTCPServer -from deepspeech.utils.socket_server import warm_up_test -from deepspeech.utils.utility import add_arguments -from deepspeech.utils.utility import print_arguments +from paddlespeech.s2t.exps.deepspeech2.config import get_cfg_defaults +from paddlespeech.s2t.io.collator import SpeechCollator +from paddlespeech.s2t.io.dataset import ManifestDataset +from paddlespeech.s2t.models.ds2 import DeepSpeech2Model +from paddlespeech.s2t.training.cli import default_argument_parser +from paddlespeech.s2t.utils.socket_server import AsrRequestHandler +from paddlespeech.s2t.utils.socket_server import AsrTCPServer +from paddlespeech.s2t.utils.socket_server import warm_up_test +from paddlespeech.s2t.utils.utility import add_arguments +from paddlespeech.s2t.utils.utility import print_arguments def start_server(config, args): diff --git a/deepspeech/exps/deepspeech2/bin/export.py b/paddlespeech/s2t/exps/deepspeech2/bin/export.py similarity index 85% rename from deepspeech/exps/deepspeech2/bin/export.py rename to paddlespeech/s2t/exps/deepspeech2/bin/export.py index 5f89e904..66042e84 100644 --- a/deepspeech/exps/deepspeech2/bin/export.py +++ b/paddlespeech/s2t/exps/deepspeech2/bin/export.py @@ -12,10 +12,10 @@ # See the License for the specific language governing permissions and # limitations under the License. """Export for DeepSpeech2 model.""" -from deepspeech.exps.deepspeech2.config import get_cfg_defaults -from deepspeech.exps.deepspeech2.model import DeepSpeech2Tester as Tester -from deepspeech.training.cli import default_argument_parser -from deepspeech.utils.utility import print_arguments +from paddlespeech.s2t.exps.deepspeech2.config import get_cfg_defaults +from paddlespeech.s2t.exps.deepspeech2.model import DeepSpeech2Tester as Tester +from paddlespeech.s2t.training.cli import default_argument_parser +from paddlespeech.s2t.utils.utility import print_arguments def main_sp(config, args): diff --git a/deepspeech/exps/deepspeech2/bin/test.py b/paddlespeech/s2t/exps/deepspeech2/bin/test.py similarity index 85% rename from deepspeech/exps/deepspeech2/bin/test.py rename to paddlespeech/s2t/exps/deepspeech2/bin/test.py index 6181ff90..f52615fa 100644 --- a/deepspeech/exps/deepspeech2/bin/test.py +++ b/paddlespeech/s2t/exps/deepspeech2/bin/test.py @@ -12,10 +12,10 @@ # See the License for the specific language governing permissions and # limitations under the License. """Evaluation for DeepSpeech2 model.""" -from deepspeech.exps.deepspeech2.config import get_cfg_defaults -from deepspeech.exps.deepspeech2.model import DeepSpeech2Tester as Tester -from deepspeech.training.cli import default_argument_parser -from deepspeech.utils.utility import print_arguments +from paddlespeech.s2t.exps.deepspeech2.config import get_cfg_defaults +from paddlespeech.s2t.exps.deepspeech2.model import DeepSpeech2Tester as Tester +from paddlespeech.s2t.training.cli import default_argument_parser +from paddlespeech.s2t.utils.utility import print_arguments def main_sp(config, args): diff --git a/deepspeech/exps/deepspeech2/bin/test_export.py b/paddlespeech/s2t/exps/deepspeech2/bin/test_export.py similarity index 85% rename from deepspeech/exps/deepspeech2/bin/test_export.py rename to paddlespeech/s2t/exps/deepspeech2/bin/test_export.py index 77bb0025..b6b34d08 100644 --- a/deepspeech/exps/deepspeech2/bin/test_export.py +++ b/paddlespeech/s2t/exps/deepspeech2/bin/test_export.py @@ -12,10 +12,10 @@ # See the License for the specific language governing permissions and # limitations under the License. """Evaluation for DeepSpeech2 model.""" -from deepspeech.exps.deepspeech2.config import get_cfg_defaults -from deepspeech.exps.deepspeech2.model import DeepSpeech2ExportTester as ExportTester -from deepspeech.training.cli import default_argument_parser -from deepspeech.utils.utility import print_arguments +from paddlespeech.s2t.exps.deepspeech2.config import get_cfg_defaults +from paddlespeech.s2t.exps.deepspeech2.model import DeepSpeech2ExportTester as ExportTester +from paddlespeech.s2t.training.cli import default_argument_parser +from paddlespeech.s2t.utils.utility import print_arguments def main_sp(config, args): diff --git a/deepspeech/exps/deepspeech2/bin/test_hub.py b/paddlespeech/s2t/exps/deepspeech2/bin/test_hub.py similarity index 90% rename from deepspeech/exps/deepspeech2/bin/test_hub.py rename to paddlespeech/s2t/exps/deepspeech2/bin/test_hub.py index 1203991d..8ab8fea2 100644 --- a/deepspeech/exps/deepspeech2/bin/test_hub.py +++ b/paddlespeech/s2t/exps/deepspeech2/bin/test_hub.py @@ -19,17 +19,17 @@ from pathlib import Path import paddle import soundfile -from deepspeech.exps.deepspeech2.config import get_cfg_defaults -from deepspeech.frontend.featurizer.text_featurizer import TextFeaturizer -from deepspeech.io.collator import SpeechCollator -from deepspeech.models.ds2 import DeepSpeech2Model -from deepspeech.models.ds2_online import DeepSpeech2ModelOnline -from deepspeech.training.cli import default_argument_parser -from deepspeech.utils import mp_tools -from deepspeech.utils.checkpoint import Checkpoint -from deepspeech.utils.log import Log -from deepspeech.utils.utility import print_arguments -from deepspeech.utils.utility import UpdateConfig +from paddlespeech.s2t.exps.deepspeech2.config import get_cfg_defaults +from paddlespeech.s2t.frontend.featurizer.text_featurizer import TextFeaturizer +from paddlespeech.s2t.io.collator import SpeechCollator +from paddlespeech.s2t.models.ds2 import DeepSpeech2Model +from paddlespeech.s2t.models.ds2_online import DeepSpeech2ModelOnline +from paddlespeech.s2t.training.cli import default_argument_parser +from paddlespeech.s2t.utils import mp_tools +from paddlespeech.s2t.utils.checkpoint import Checkpoint +from paddlespeech.s2t.utils.log import Log +from paddlespeech.s2t.utils.utility import print_arguments +from paddlespeech.s2t.utils.utility import UpdateConfig logger = Log(__name__).getlog() diff --git a/deepspeech/exps/deepspeech2/bin/train.py b/paddlespeech/s2t/exps/deepspeech2/bin/train.py similarity index 85% rename from deepspeech/exps/deepspeech2/bin/train.py rename to paddlespeech/s2t/exps/deepspeech2/bin/train.py index 02aefe3d..d9b610a0 100644 --- a/deepspeech/exps/deepspeech2/bin/train.py +++ b/paddlespeech/s2t/exps/deepspeech2/bin/train.py @@ -14,10 +14,10 @@ """Trainer for DeepSpeech2 model.""" from paddle import distributed as dist -from deepspeech.exps.deepspeech2.config import get_cfg_defaults -from deepspeech.exps.deepspeech2.model import DeepSpeech2Trainer as Trainer -from deepspeech.training.cli import default_argument_parser -from deepspeech.utils.utility import print_arguments +from paddlespeech.s2t.exps.deepspeech2.config import get_cfg_defaults +from paddlespeech.s2t.exps.deepspeech2.model import DeepSpeech2Trainer as Trainer +from paddlespeech.s2t.training.cli import default_argument_parser +from paddlespeech.s2t.utils.utility import print_arguments def main_sp(config, args): diff --git a/deepspeech/exps/deepspeech2/config.py b/paddlespeech/s2t/exps/deepspeech2/config.py similarity index 77% rename from deepspeech/exps/deepspeech2/config.py rename to paddlespeech/s2t/exps/deepspeech2/config.py index 38b7d0e4..58dc05ff 100644 --- a/deepspeech/exps/deepspeech2/config.py +++ b/paddlespeech/s2t/exps/deepspeech2/config.py @@ -13,12 +13,12 @@ # limitations under the License. from yacs.config import CfgNode -from deepspeech.exps.deepspeech2.model import DeepSpeech2Tester -from deepspeech.exps.deepspeech2.model import DeepSpeech2Trainer -from deepspeech.io.collator import SpeechCollator -from deepspeech.io.dataset import ManifestDataset -from deepspeech.models.ds2 import DeepSpeech2Model -from deepspeech.models.ds2_online import DeepSpeech2ModelOnline +from paddlespeech.s2t.exps.deepspeech2.model import DeepSpeech2Tester +from paddlespeech.s2t.exps.deepspeech2.model import DeepSpeech2Trainer +from paddlespeech.s2t.io.collator import SpeechCollator +from paddlespeech.s2t.io.dataset import ManifestDataset +from paddlespeech.s2t.models.ds2 import DeepSpeech2Model +from paddlespeech.s2t.models.ds2_online import DeepSpeech2ModelOnline def get_cfg_defaults(model_type='offline'): diff --git a/deepspeech/exps/deepspeech2/model.py b/paddlespeech/s2t/exps/deepspeech2/model.py similarity index 95% rename from deepspeech/exps/deepspeech2/model.py rename to paddlespeech/s2t/exps/deepspeech2/model.py index 15281177..683fab14 100644 --- a/deepspeech/exps/deepspeech2/model.py +++ b/paddlespeech/s2t/exps/deepspeech2/model.py @@ -26,25 +26,25 @@ from paddle import inference from paddle.io import DataLoader from yacs.config import CfgNode -from deepspeech.frontend.featurizer.text_featurizer import TextFeaturizer -from deepspeech.io.collator import SpeechCollator -from deepspeech.io.dataset import ManifestDataset -from deepspeech.io.sampler import SortagradBatchSampler -from deepspeech.io.sampler import SortagradDistributedBatchSampler -from deepspeech.models.ds2 import DeepSpeech2InferModel -from deepspeech.models.ds2 import DeepSpeech2Model -from deepspeech.models.ds2_online import DeepSpeech2InferModelOnline -from deepspeech.models.ds2_online import DeepSpeech2ModelOnline -from deepspeech.training.gradclip import ClipGradByGlobalNormWithLog -from deepspeech.training.reporter import report -from deepspeech.training.timer import Timer -from deepspeech.training.trainer import Trainer -from deepspeech.utils import error_rate -from deepspeech.utils import layer_tools -from deepspeech.utils import mp_tools -from deepspeech.utils.log import Autolog -from deepspeech.utils.log import Log -from deepspeech.utils.utility import UpdateConfig +from paddlespeech.s2t.frontend.featurizer.text_featurizer import TextFeaturizer +from paddlespeech.s2t.io.collator import SpeechCollator +from paddlespeech.s2t.io.dataset import ManifestDataset +from paddlespeech.s2t.io.sampler import SortagradBatchSampler +from paddlespeech.s2t.io.sampler import SortagradDistributedBatchSampler +from paddlespeech.s2t.models.ds2 import DeepSpeech2InferModel +from paddlespeech.s2t.models.ds2 import DeepSpeech2Model +from paddlespeech.s2t.models.ds2_online import DeepSpeech2InferModelOnline +from paddlespeech.s2t.models.ds2_online import DeepSpeech2ModelOnline +from paddlespeech.s2t.training.gradclip import ClipGradByGlobalNormWithLog +from paddlespeech.s2t.training.reporter import report +from paddlespeech.s2t.training.timer import Timer +from paddlespeech.s2t.training.trainer import Trainer +from paddlespeech.s2t.utils import error_rate +from paddlespeech.s2t.utils import layer_tools +from paddlespeech.s2t.utils import mp_tools +from paddlespeech.s2t.utils.log import Autolog +from paddlespeech.s2t.utils.log import Log +from paddlespeech.s2t.utils.utility import UpdateConfig logger = Log(__name__).getlog() diff --git a/deepspeech/exps/lm/transformer/__init__.py b/paddlespeech/s2t/exps/lm/transformer/__init__.py similarity index 100% rename from deepspeech/exps/lm/transformer/__init__.py rename to paddlespeech/s2t/exps/lm/transformer/__init__.py diff --git a/deepspeech/exps/lm/transformer/bin/cacu_perplexity.py b/paddlespeech/s2t/exps/lm/transformer/bin/cacu_perplexity.py similarity index 96% rename from deepspeech/exps/lm/transformer/bin/cacu_perplexity.py rename to paddlespeech/s2t/exps/lm/transformer/bin/cacu_perplexity.py index 29a880f8..f3e4d209 100644 --- a/deepspeech/exps/lm/transformer/bin/cacu_perplexity.py +++ b/paddlespeech/s2t/exps/lm/transformer/bin/cacu_perplexity.py @@ -74,7 +74,7 @@ def get_parser(): def main(args): parser = get_parser() args = parser.parse_args(args) - from deepspeech.exps.lm.transformer.lm_cacu_perplexity import run_get_perplexity + from paddlespeech.s2t.exps.lm.transformer.lm_cacu_perplexity import run_get_perplexity run_get_perplexity(args) diff --git a/deepspeech/exps/lm/transformer/lm_cacu_perplexity.py b/paddlespeech/s2t/exps/lm/transformer/lm_cacu_perplexity.py similarity index 94% rename from deepspeech/exps/lm/transformer/lm_cacu_perplexity.py rename to paddlespeech/s2t/exps/lm/transformer/lm_cacu_perplexity.py index b63bcd08..ab0ec8f0 100644 --- a/deepspeech/exps/lm/transformer/lm_cacu_perplexity.py +++ b/paddlespeech/s2t/exps/lm/transformer/lm_cacu_perplexity.py @@ -19,10 +19,10 @@ import paddle from paddle.io import DataLoader from yacs.config import CfgNode -from deepspeech.io.collator import TextCollatorSpm -from deepspeech.io.dataset import TextDataset -from deepspeech.models.lm_interface import dynamic_import_lm -from deepspeech.utils.log import Log +from paddlespeech.s2t.io.collator import TextCollatorSpm +from paddlespeech.s2t.io.dataset import TextDataset +from paddlespeech.s2t.models.lm_interface import dynamic_import_lm +from paddlespeech.s2t.utils.log import Log logger = Log(__name__).getlog() diff --git a/deepspeech/exps/u2/__init__.py b/paddlespeech/s2t/exps/u2/__init__.py similarity index 100% rename from deepspeech/exps/u2/__init__.py rename to paddlespeech/s2t/exps/u2/__init__.py diff --git a/deepspeech/exps/u2/bin/alignment.py b/paddlespeech/s2t/exps/u2/bin/alignment.py similarity index 85% rename from deepspeech/exps/u2/bin/alignment.py rename to paddlespeech/s2t/exps/u2/bin/alignment.py index 91aac1a8..df95baeb 100644 --- a/deepspeech/exps/u2/bin/alignment.py +++ b/paddlespeech/s2t/exps/u2/bin/alignment.py @@ -12,10 +12,10 @@ # See the License for the specific language governing permissions and # limitations under the License. """Alignment for U2 model.""" -from deepspeech.exps.u2.config import get_cfg_defaults -from deepspeech.exps.u2.model import U2Tester as Tester -from deepspeech.training.cli import default_argument_parser -from deepspeech.utils.utility import print_arguments +from paddlespeech.s2t.exps.u2.config import get_cfg_defaults +from paddlespeech.s2t.exps.u2.model import U2Tester as Tester +from paddlespeech.s2t.training.cli import default_argument_parser +from paddlespeech.s2t.utils.utility import print_arguments def main_sp(config, args): diff --git a/deepspeech/exps/u2/bin/export.py b/paddlespeech/s2t/exps/u2/bin/export.py similarity index 85% rename from deepspeech/exps/u2/bin/export.py rename to paddlespeech/s2t/exps/u2/bin/export.py index 89acbf70..44fc7c3e 100644 --- a/deepspeech/exps/u2/bin/export.py +++ b/paddlespeech/s2t/exps/u2/bin/export.py @@ -12,10 +12,10 @@ # See the License for the specific language governing permissions and # limitations under the License. """Export for U2 model.""" -from deepspeech.exps.u2.config import get_cfg_defaults -from deepspeech.exps.u2.model import U2Tester as Tester -from deepspeech.training.cli import default_argument_parser -from deepspeech.utils.utility import print_arguments +from paddlespeech.s2t.exps.u2.config import get_cfg_defaults +from paddlespeech.s2t.exps.u2.model import U2Tester as Tester +from paddlespeech.s2t.training.cli import default_argument_parser +from paddlespeech.s2t.utils.utility import print_arguments def main_sp(config, args): diff --git a/deepspeech/exps/u2/bin/test.py b/paddlespeech/s2t/exps/u2/bin/test.py similarity index 86% rename from deepspeech/exps/u2/bin/test.py rename to paddlespeech/s2t/exps/u2/bin/test.py index 3068f1c7..48b0670d 100644 --- a/deepspeech/exps/u2/bin/test.py +++ b/paddlespeech/s2t/exps/u2/bin/test.py @@ -14,10 +14,10 @@ """Evaluation for U2 model.""" import cProfile -from deepspeech.exps.u2.config import get_cfg_defaults -from deepspeech.exps.u2.model import U2Tester as Tester -from deepspeech.training.cli import default_argument_parser -from deepspeech.utils.utility import print_arguments +from paddlespeech.s2t.exps.u2.config import get_cfg_defaults +from paddlespeech.s2t.exps.u2.model import U2Tester as Tester +from paddlespeech.s2t.training.cli import default_argument_parser +from paddlespeech.s2t.utils.utility import print_arguments # TODO(hui zhang): dynamic load diff --git a/deepspeech/exps/u2/bin/test_hub.py b/paddlespeech/s2t/exps/u2/bin/test_hub.py similarity index 90% rename from deepspeech/exps/u2/bin/test_hub.py rename to paddlespeech/s2t/exps/u2/bin/test_hub.py index 8c006ebe..55a61d5c 100644 --- a/deepspeech/exps/u2/bin/test_hub.py +++ b/paddlespeech/s2t/exps/u2/bin/test_hub.py @@ -19,17 +19,17 @@ import sys import paddle import soundfile -from deepspeech.exps.u2.config import get_cfg_defaults -from deepspeech.frontend.featurizer.text_featurizer import TextFeaturizer -from deepspeech.io.collator import SpeechCollator -from deepspeech.models.u2 import U2Model -from deepspeech.training.cli import default_argument_parser -from deepspeech.training.trainer import Trainer -from deepspeech.utils import layer_tools -from deepspeech.utils import mp_tools -from deepspeech.utils.log import Log -from deepspeech.utils.utility import print_arguments -from deepspeech.utils.utility import UpdateConfig +from paddlespeech.s2t.exps.u2.config import get_cfg_defaults +from paddlespeech.s2t.frontend.featurizer.text_featurizer import TextFeaturizer +from paddlespeech.s2t.io.collator import SpeechCollator +from paddlespeech.s2t.models.u2 import U2Model +from paddlespeech.s2t.training.cli import default_argument_parser +from paddlespeech.s2t.training.trainer import Trainer +from paddlespeech.s2t.utils import layer_tools +from paddlespeech.s2t.utils import mp_tools +from paddlespeech.s2t.utils.log import Log +from paddlespeech.s2t.utils.utility import print_arguments +from paddlespeech.s2t.utils.utility import UpdateConfig logger = Log(__name__).getlog() # TODO(hui zhang): dynamic load diff --git a/deepspeech/exps/u2/bin/train.py b/paddlespeech/s2t/exps/u2/bin/train.py similarity index 83% rename from deepspeech/exps/u2/bin/train.py rename to paddlespeech/s2t/exps/u2/bin/train.py index 17fb08a6..127db521 100644 --- a/deepspeech/exps/u2/bin/train.py +++ b/paddlespeech/s2t/exps/u2/bin/train.py @@ -17,12 +17,12 @@ import os from paddle import distributed as dist -from deepspeech.exps.u2.config import get_cfg_defaults -from deepspeech.exps.u2.model import U2Trainer as Trainer -from deepspeech.training.cli import default_argument_parser -from deepspeech.utils.utility import print_arguments +from paddlespeech.s2t.exps.u2.config import get_cfg_defaults +from paddlespeech.s2t.exps.u2.model import U2Trainer as Trainer +from paddlespeech.s2t.training.cli import default_argument_parser +from paddlespeech.s2t.utils.utility import print_arguments -# from deepspeech.exps.u2.trainer import U2Trainer as Trainer +# from paddlespeech.s2t.exps.u2.trainer import U2Trainer as Trainer def main_sp(config, args): diff --git a/deepspeech/exps/u2/config.py b/paddlespeech/s2t/exps/u2/config.py similarity index 80% rename from deepspeech/exps/u2/config.py rename to paddlespeech/s2t/exps/u2/config.py index 4ec7bd19..898b0bb2 100644 --- a/deepspeech/exps/u2/config.py +++ b/paddlespeech/s2t/exps/u2/config.py @@ -13,11 +13,11 @@ # limitations under the License. from yacs.config import CfgNode -from deepspeech.exps.u2.model import U2Tester -from deepspeech.exps.u2.model import U2Trainer -from deepspeech.io.collator import SpeechCollator -from deepspeech.io.dataset import ManifestDataset -from deepspeech.models.u2 import U2Model +from paddlespeech.s2t.exps.u2.model import U2Tester +from paddlespeech.s2t.exps.u2.model import U2Trainer +from paddlespeech.s2t.io.collator import SpeechCollator +from paddlespeech.s2t.io.dataset import ManifestDataset +from paddlespeech.s2t.models.u2 import U2Model _C = CfgNode() diff --git a/deepspeech/exps/u2/model.py b/paddlespeech/s2t/exps/u2/model.py similarity index 96% rename from deepspeech/exps/u2/model.py rename to paddlespeech/s2t/exps/u2/model.py index e47a59ed..67441fae 100644 --- a/deepspeech/exps/u2/model.py +++ b/paddlespeech/s2t/exps/u2/model.py @@ -27,23 +27,23 @@ from paddle import distributed as dist from paddle.io import DataLoader from yacs.config import CfgNode -from deepspeech.io.collator import SpeechCollator -from deepspeech.io.dataset import ManifestDataset -from deepspeech.io.sampler import SortagradBatchSampler -from deepspeech.io.sampler import SortagradDistributedBatchSampler -from deepspeech.models.u2 import U2Model -from deepspeech.training.optimizer import OptimizerFactory -from deepspeech.training.reporter import ObsScope -from deepspeech.training.reporter import report -from deepspeech.training.scheduler import LRSchedulerFactory -from deepspeech.training.timer import Timer -from deepspeech.training.trainer import Trainer -from deepspeech.utils import ctc_utils -from deepspeech.utils import error_rate -from deepspeech.utils import layer_tools -from deepspeech.utils import mp_tools -from deepspeech.utils.log import Log -from deepspeech.utils.utility import UpdateConfig +from paddlespeech.s2t.io.collator import SpeechCollator +from paddlespeech.s2t.io.dataset import ManifestDataset +from paddlespeech.s2t.io.sampler import SortagradBatchSampler +from paddlespeech.s2t.io.sampler import SortagradDistributedBatchSampler +from paddlespeech.s2t.models.u2 import U2Model +from paddlespeech.s2t.training.optimizer import OptimizerFactory +from paddlespeech.s2t.training.reporter import ObsScope +from paddlespeech.s2t.training.reporter import report +from paddlespeech.s2t.training.scheduler import LRSchedulerFactory +from paddlespeech.s2t.training.timer import Timer +from paddlespeech.s2t.training.trainer import Trainer +from paddlespeech.s2t.utils import ctc_utils +from paddlespeech.s2t.utils import error_rate +from paddlespeech.s2t.utils import layer_tools +from paddlespeech.s2t.utils import mp_tools +from paddlespeech.s2t.utils.log import Log +from paddlespeech.s2t.utils.utility import UpdateConfig logger = Log(__name__).getlog() @@ -566,7 +566,7 @@ class U2Tester(U2Trainer): nn.Layer: inference model List[paddle.static.InputSpec]: input spec. """ - from deepspeech.models.u2 import U2InferModel + from paddlespeech.s2t.models.u2 import U2InferModel infer_model = U2InferModel.from_pretrained(self.test_loader, self.config.model.clone(), self.args.checkpoint_path) diff --git a/deepspeech/exps/u2/trainer.py b/paddlespeech/s2t/exps/u2/trainer.py similarity index 88% rename from deepspeech/exps/u2/trainer.py rename to paddlespeech/s2t/exps/u2/trainer.py index 1cb3eccd..22a0a3c5 100644 --- a/deepspeech/exps/u2/trainer.py +++ b/paddlespeech/s2t/exps/u2/trainer.py @@ -16,23 +16,23 @@ import paddle from paddle import distributed as dist from paddle.io import DataLoader -from deepspeech.io.collator import SpeechCollator -from deepspeech.io.dataset import ManifestDataset -from deepspeech.io.sampler import SortagradBatchSampler -from deepspeech.io.sampler import SortagradDistributedBatchSampler -from deepspeech.models.u2 import U2Evaluator -from deepspeech.models.u2 import U2Model -from deepspeech.models.u2 import U2Updater -from deepspeech.training.extensions.snapshot import Snapshot -from deepspeech.training.extensions.visualizer import VisualDL -from deepspeech.training.optimizer import OptimizerFactory -from deepspeech.training.scheduler import LRSchedulerFactory -from deepspeech.training.timer import Timer -from deepspeech.training.trainer import Trainer -from deepspeech.training.updaters.trainer import Trainer as NewTrainer -from deepspeech.utils import layer_tools -from deepspeech.utils.log import Log -from deepspeech.utils.utility import UpdateConfig +from paddlespeech.s2t.io.collator import SpeechCollator +from paddlespeech.s2t.io.dataset import ManifestDataset +from paddlespeech.s2t.io.sampler import SortagradBatchSampler +from paddlespeech.s2t.io.sampler import SortagradDistributedBatchSampler +from paddlespeech.s2t.models.u2 import U2Evaluator +from paddlespeech.s2t.models.u2 import U2Model +from paddlespeech.s2t.models.u2 import U2Updater +from paddlespeech.s2t.training.extensions.snapshot import Snapshot +from paddlespeech.s2t.training.extensions.visualizer import VisualDL +from paddlespeech.s2t.training.optimizer import OptimizerFactory +from paddlespeech.s2t.training.scheduler import LRSchedulerFactory +from paddlespeech.s2t.training.timer import Timer +from paddlespeech.s2t.training.trainer import Trainer +from paddlespeech.s2t.training.updaters.trainer import Trainer as NewTrainer +from paddlespeech.s2t.utils import layer_tools +from paddlespeech.s2t.utils.log import Log +from paddlespeech.s2t.utils.utility import UpdateConfig logger = Log(__name__).getlog() diff --git a/deepspeech/exps/u2_kaldi/__init__.py b/paddlespeech/s2t/exps/u2_kaldi/__init__.py similarity index 100% rename from deepspeech/exps/u2_kaldi/__init__.py rename to paddlespeech/s2t/exps/u2_kaldi/__init__.py diff --git a/deepspeech/exps/u2_kaldi/bin/recog.py b/paddlespeech/s2t/exps/u2_kaldi/bin/recog.py similarity index 92% rename from deepspeech/exps/u2_kaldi/bin/recog.py rename to paddlespeech/s2t/exps/u2_kaldi/bin/recog.py index e94a1ab1..37ddd229 100644 --- a/deepspeech/exps/u2_kaldi/bin/recog.py +++ b/paddlespeech/s2t/exps/u2_kaldi/bin/recog.py @@ -13,7 +13,7 @@ # limitations under the License. import sys -from deepspeech.decoders.recog_bin import main +from paddlespeech.s2t.decoders.recog_bin import main if __name__ == "__main__": main(sys.argv[1:]) diff --git a/deepspeech/exps/u2_kaldi/bin/test.py b/paddlespeech/s2t/exps/u2_kaldi/bin/test.py similarity index 88% rename from deepspeech/exps/u2_kaldi/bin/test.py rename to paddlespeech/s2t/exps/u2_kaldi/bin/test.py index eb7b0280..67bed349 100644 --- a/deepspeech/exps/u2_kaldi/bin/test.py +++ b/paddlespeech/s2t/exps/u2_kaldi/bin/test.py @@ -16,13 +16,13 @@ import cProfile from yacs.config import CfgNode -from deepspeech.training.cli import default_argument_parser -from deepspeech.utils.dynamic_import import dynamic_import -from deepspeech.utils.utility import print_arguments +from paddlespeech.s2t.training.cli import default_argument_parser +from paddlespeech.s2t.utils.dynamic_import import dynamic_import +from paddlespeech.s2t.utils.utility import print_arguments model_test_alias = { - "u2": "deepspeech.exps.u2.model:U2Tester", - "u2_kaldi": "deepspeech.exps.u2_kaldi.model:U2Tester", + "u2": "paddlespeech.s2t.exps.u2.model:U2Tester", + "u2_kaldi": "paddlespeech.s2t.exps.u2_kaldi.model:U2Tester", } diff --git a/deepspeech/exps/u2_kaldi/bin/train.py b/paddlespeech/s2t/exps/u2_kaldi/bin/train.py similarity index 85% rename from deepspeech/exps/u2_kaldi/bin/train.py rename to paddlespeech/s2t/exps/u2_kaldi/bin/train.py index d909727f..d3427eec 100644 --- a/deepspeech/exps/u2_kaldi/bin/train.py +++ b/paddlespeech/s2t/exps/u2_kaldi/bin/train.py @@ -18,13 +18,13 @@ import os from paddle import distributed as dist from yacs.config import CfgNode -from deepspeech.training.cli import default_argument_parser -from deepspeech.utils.dynamic_import import dynamic_import -from deepspeech.utils.utility import print_arguments +from paddlespeech.s2t.training.cli import default_argument_parser +from paddlespeech.s2t.utils.dynamic_import import dynamic_import +from paddlespeech.s2t.utils.utility import print_arguments model_train_alias = { - "u2": "deepspeech.exps.u2.model:U2Trainer", - "u2_kaldi": "deepspeech.exps.u2_kaldi.model:U2Trainer", + "u2": "paddlespeech.s2t.exps.u2.model:U2Trainer", + "u2_kaldi": "paddlespeech.s2t.exps.u2_kaldi.model:U2Trainer", } diff --git a/deepspeech/exps/u2_kaldi/model.py b/paddlespeech/s2t/exps/u2_kaldi/model.py similarity index 96% rename from deepspeech/exps/u2_kaldi/model.py rename to paddlespeech/s2t/exps/u2_kaldi/model.py index 663c36d8..e37784aa 100644 --- a/deepspeech/exps/u2_kaldi/model.py +++ b/paddlespeech/s2t/exps/u2_kaldi/model.py @@ -25,20 +25,20 @@ import paddle from paddle import distributed as dist from yacs.config import CfgNode -from deepspeech.frontend.featurizer import TextFeaturizer -from deepspeech.frontend.utility import load_dict -from deepspeech.io.dataloader import BatchDataLoader -from deepspeech.models.u2 import U2Model -from deepspeech.training.optimizer import OptimizerFactory -from deepspeech.training.scheduler import LRSchedulerFactory -from deepspeech.training.timer import Timer -from deepspeech.training.trainer import Trainer -from deepspeech.utils import ctc_utils -from deepspeech.utils import error_rate -from deepspeech.utils import layer_tools -from deepspeech.utils import mp_tools -from deepspeech.utils.log import Log -from deepspeech.utils.utility import UpdateConfig +from paddlespeech.s2t.frontend.featurizer import TextFeaturizer +from paddlespeech.s2t.frontend.utility import load_dict +from paddlespeech.s2t.io.dataloader import BatchDataLoader +from paddlespeech.s2t.models.u2 import U2Model +from paddlespeech.s2t.training.optimizer import OptimizerFactory +from paddlespeech.s2t.training.scheduler import LRSchedulerFactory +from paddlespeech.s2t.training.timer import Timer +from paddlespeech.s2t.training.trainer import Trainer +from paddlespeech.s2t.utils import ctc_utils +from paddlespeech.s2t.utils import error_rate +from paddlespeech.s2t.utils import layer_tools +from paddlespeech.s2t.utils import mp_tools +from paddlespeech.s2t.utils.log import Log +from paddlespeech.s2t.utils.utility import UpdateConfig logger = Log(__name__).getlog() @@ -539,7 +539,7 @@ class U2Tester(U2Trainer): nn.Layer: inference model List[paddle.static.InputSpec]: input spec. """ - from deepspeech.models.u2 import U2InferModel + from paddlespeech.s2t.models.u2 import U2InferModel infer_model = U2InferModel.from_pretrained(self.test_loader, self.config.model.clone(), self.args.checkpoint_path) diff --git a/deepspeech/exps/u2_st/__init__.py b/paddlespeech/s2t/exps/u2_st/__init__.py similarity index 100% rename from deepspeech/exps/u2_st/__init__.py rename to paddlespeech/s2t/exps/u2_st/__init__.py diff --git a/deepspeech/exps/u2_st/bin/export.py b/paddlespeech/s2t/exps/u2_st/bin/export.py similarity index 84% rename from deepspeech/exps/u2_st/bin/export.py rename to paddlespeech/s2t/exps/u2_st/bin/export.py index 2486f4c2..69d9718f 100644 --- a/deepspeech/exps/u2_st/bin/export.py +++ b/paddlespeech/s2t/exps/u2_st/bin/export.py @@ -12,10 +12,10 @@ # See the License for the specific language governing permissions and # limitations under the License. """Export for U2 model.""" -from deepspeech.exps.u2_st.config import get_cfg_defaults -from deepspeech.exps.u2_st.model import U2STTester as Tester -from deepspeech.training.cli import default_argument_parser -from deepspeech.utils.utility import print_arguments +from paddlespeech.s2t.exps.u2_st.config import get_cfg_defaults +from paddlespeech.s2t.exps.u2_st.model import U2STTester as Tester +from paddlespeech.s2t.training.cli import default_argument_parser +from paddlespeech.s2t.utils.utility import print_arguments def main_sp(config, args): diff --git a/deepspeech/exps/u2_st/bin/test.py b/paddlespeech/s2t/exps/u2_st/bin/test.py similarity index 86% rename from deepspeech/exps/u2_st/bin/test.py rename to paddlespeech/s2t/exps/u2_st/bin/test.py index 3d51b60e..93c2fee0 100644 --- a/deepspeech/exps/u2_st/bin/test.py +++ b/paddlespeech/s2t/exps/u2_st/bin/test.py @@ -14,10 +14,10 @@ """Evaluation for U2 model.""" import cProfile -from deepspeech.exps.u2_st.config import get_cfg_defaults -from deepspeech.exps.u2_st.model import U2STTester as Tester -from deepspeech.training.cli import default_argument_parser -from deepspeech.utils.utility import print_arguments +from paddlespeech.s2t.exps.u2_st.config import get_cfg_defaults +from paddlespeech.s2t.exps.u2_st.model import U2STTester as Tester +from paddlespeech.s2t.training.cli import default_argument_parser +from paddlespeech.s2t.utils.utility import print_arguments # TODO(hui zhang): dynamic load diff --git a/deepspeech/exps/u2_st/bin/train.py b/paddlespeech/s2t/exps/u2_st/bin/train.py similarity index 86% rename from deepspeech/exps/u2_st/bin/train.py rename to paddlespeech/s2t/exps/u2_st/bin/train.py index 1e6a746b..3d823cc4 100644 --- a/deepspeech/exps/u2_st/bin/train.py +++ b/paddlespeech/s2t/exps/u2_st/bin/train.py @@ -17,10 +17,10 @@ import os from paddle import distributed as dist -from deepspeech.exps.u2_st.config import get_cfg_defaults -from deepspeech.exps.u2_st.model import U2STTrainer as Trainer -from deepspeech.training.cli import default_argument_parser -from deepspeech.utils.utility import print_arguments +from paddlespeech.s2t.exps.u2_st.config import get_cfg_defaults +from paddlespeech.s2t.exps.u2_st.model import U2STTrainer as Trainer +from paddlespeech.s2t.training.cli import default_argument_parser +from paddlespeech.s2t.utils.utility import print_arguments def main_sp(config, args): diff --git a/deepspeech/exps/u2_st/config.py b/paddlespeech/s2t/exps/u2_st/config.py similarity index 80% rename from deepspeech/exps/u2_st/config.py rename to paddlespeech/s2t/exps/u2_st/config.py index 818b0e25..a48f9106 100644 --- a/deepspeech/exps/u2_st/config.py +++ b/paddlespeech/s2t/exps/u2_st/config.py @@ -13,11 +13,11 @@ # limitations under the License. from yacs.config import CfgNode -from deepspeech.exps.u2_st.model import U2STTester -from deepspeech.exps.u2_st.model import U2STTrainer -from deepspeech.io.collator import SpeechCollator -from deepspeech.io.dataset import ManifestDataset -from deepspeech.models.u2_st import U2STModel +from paddlespeech.s2t.exps.u2_st.model import U2STTester +from paddlespeech.s2t.exps.u2_st.model import U2STTrainer +from paddlespeech.s2t.io.collator import SpeechCollator +from paddlespeech.s2t.io.dataset import ManifestDataset +from paddlespeech.s2t.models.u2_st import U2STModel _C = CfgNode() diff --git a/deepspeech/exps/u2_st/model.py b/paddlespeech/s2t/exps/u2_st/model.py similarity index 96% rename from deepspeech/exps/u2_st/model.py rename to paddlespeech/s2t/exps/u2_st/model.py index 1f638e64..f458216e 100644 --- a/deepspeech/exps/u2_st/model.py +++ b/paddlespeech/s2t/exps/u2_st/model.py @@ -26,22 +26,22 @@ from paddle import distributed as dist from paddle.io import DataLoader from yacs.config import CfgNode -from deepspeech.io.collator import SpeechCollator -from deepspeech.io.collator import TripletSpeechCollator -from deepspeech.io.dataset import ManifestDataset -from deepspeech.io.sampler import SortagradBatchSampler -from deepspeech.io.sampler import SortagradDistributedBatchSampler -from deepspeech.models.u2_st import U2STModel -from deepspeech.training.gradclip import ClipGradByGlobalNormWithLog -from deepspeech.training.scheduler import WarmupLR -from deepspeech.training.timer import Timer -from deepspeech.training.trainer import Trainer -from deepspeech.utils import bleu_score -from deepspeech.utils import ctc_utils -from deepspeech.utils import layer_tools -from deepspeech.utils import mp_tools -from deepspeech.utils.log import Log -from deepspeech.utils.utility import UpdateConfig +from paddlespeech.s2t.io.collator import SpeechCollator +from paddlespeech.s2t.io.collator import TripletSpeechCollator +from paddlespeech.s2t.io.dataset import ManifestDataset +from paddlespeech.s2t.io.sampler import SortagradBatchSampler +from paddlespeech.s2t.io.sampler import SortagradDistributedBatchSampler +from paddlespeech.s2t.models.u2_st import U2STModel +from paddlespeech.s2t.training.gradclip import ClipGradByGlobalNormWithLog +from paddlespeech.s2t.training.scheduler import WarmupLR +from paddlespeech.s2t.training.timer import Timer +from paddlespeech.s2t.training.trainer import Trainer +from paddlespeech.s2t.utils import bleu_score +from paddlespeech.s2t.utils import ctc_utils +from paddlespeech.s2t.utils import layer_tools +from paddlespeech.s2t.utils import mp_tools +from paddlespeech.s2t.utils.log import Log +from paddlespeech.s2t.utils.utility import UpdateConfig logger = Log(__name__).getlog() @@ -555,7 +555,7 @@ class U2STTester(U2STTrainer): nn.Layer: inference model List[paddle.static.InputSpec]: input spec. """ - from deepspeech.models.u2 import U2InferModel + from paddlespeech.s2t.models.u2 import U2InferModel infer_model = U2InferModel.from_pretrained(self.test_loader, self.config.model.clone(), self.args.checkpoint_path) diff --git a/deepspeech/frontend/__init__.py b/paddlespeech/s2t/frontend/__init__.py similarity index 100% rename from deepspeech/frontend/__init__.py rename to paddlespeech/s2t/frontend/__init__.py diff --git a/deepspeech/frontend/audio.py b/paddlespeech/s2t/frontend/audio.py similarity index 100% rename from deepspeech/frontend/audio.py rename to paddlespeech/s2t/frontend/audio.py diff --git a/deepspeech/frontend/augmentor/__init__.py b/paddlespeech/s2t/frontend/augmentor/__init__.py similarity index 100% rename from deepspeech/frontend/augmentor/__init__.py rename to paddlespeech/s2t/frontend/augmentor/__init__.py diff --git a/deepspeech/frontend/augmentor/augmentation.py b/paddlespeech/s2t/frontend/augmentor/augmentation.py similarity index 89% rename from deepspeech/frontend/augmentor/augmentation.py rename to paddlespeech/s2t/frontend/augmentor/augmentation.py index d2316ab1..4c5ca4fe 100644 --- a/deepspeech/frontend/augmentor/augmentation.py +++ b/paddlespeech/s2t/frontend/augmentor/augmentation.py @@ -20,23 +20,23 @@ from pprint import pformat import numpy as np -from deepspeech.frontend.augmentor.base import AugmentorBase -from deepspeech.utils.dynamic_import import dynamic_import -from deepspeech.utils.log import Log +from paddlespeech.s2t.frontend.augmentor.base import AugmentorBase +from paddlespeech.s2t.utils.dynamic_import import dynamic_import +from paddlespeech.s2t.utils.log import Log logger = Log(__name__).getlog() __all__ = ["AugmentationPipeline"] import_alias = dict( - volume="deepspeech.frontend.augmentor.impulse_response:VolumePerturbAugmentor", - shift="deepspeech.frontend.augmentor.shift_perturb:ShiftPerturbAugmentor", - speed="deepspeech.frontend.augmentor.speed_perturb:SpeedPerturbAugmentor", - resample="deepspeech.frontend.augmentor.resample:ResampleAugmentor", - bayesian_normal="deepspeech.frontend.augmentor.online_bayesian_normalization:OnlineBayesianNormalizationAugmentor", - noise="deepspeech.frontend.augmentor.noise_perturb:NoisePerturbAugmentor", - impulse="deepspeech.frontend.augmentor.impulse_response:ImpulseResponseAugmentor", - specaug="deepspeech.frontend.augmentor.spec_augment:SpecAugmentor", ) + volume="paddlespeech.s2t.frontend.augmentor.impulse_response:VolumePerturbAugmentor", + shift="paddlespeech.s2t.frontend.augmentor.shift_perturb:ShiftPerturbAugmentor", + speed="paddlespeech.s2t.frontend.augmentor.speed_perturb:SpeedPerturbAugmentor", + resample="paddlespeech.s2t.frontend.augmentor.resample:ResampleAugmentor", + bayesian_normal="paddlespeech.s2t.frontend.augmentor.online_bayesian_normalization:OnlineBayesianNormalizationAugmentor", + noise="paddlespeech.s2t.frontend.augmentor.noise_perturb:NoisePerturbAugmentor", + impulse="paddlespeech.s2t.frontend.augmentor.impulse_response:ImpulseResponseAugmentor", + specaug="paddlespeech.s2t.frontend.augmentor.spec_augment:SpecAugmentor", ) class AugmentationPipeline(): diff --git a/deepspeech/frontend/augmentor/base.py b/paddlespeech/s2t/frontend/augmentor/base.py similarity index 100% rename from deepspeech/frontend/augmentor/base.py rename to paddlespeech/s2t/frontend/augmentor/base.py diff --git a/deepspeech/frontend/augmentor/impulse_response.py b/paddlespeech/s2t/frontend/augmentor/impulse_response.py similarity index 90% rename from deepspeech/frontend/augmentor/impulse_response.py rename to paddlespeech/s2t/frontend/augmentor/impulse_response.py index 818251ed..6cc9c0d4 100644 --- a/deepspeech/frontend/augmentor/impulse_response.py +++ b/paddlespeech/s2t/frontend/augmentor/impulse_response.py @@ -12,9 +12,9 @@ # See the License for the specific language governing permissions and # limitations under the License. """Contains the impulse response augmentation model.""" -from deepspeech.frontend.audio import AudioSegment -from deepspeech.frontend.augmentor.base import AugmentorBase -from deepspeech.frontend.utility import read_manifest +from paddlespeech.s2t.frontend.audio import AudioSegment +from paddlespeech.s2t.frontend.augmentor.base import AugmentorBase +from paddlespeech.s2t.frontend.utility import read_manifest class ImpulseResponseAugmentor(AugmentorBase): diff --git a/deepspeech/frontend/augmentor/noise_perturb.py b/paddlespeech/s2t/frontend/augmentor/noise_perturb.py similarity index 93% rename from deepspeech/frontend/augmentor/noise_perturb.py rename to paddlespeech/s2t/frontend/augmentor/noise_perturb.py index 790b0c39..9d6da1a8 100644 --- a/deepspeech/frontend/augmentor/noise_perturb.py +++ b/paddlespeech/s2t/frontend/augmentor/noise_perturb.py @@ -12,9 +12,9 @@ # See the License for the specific language governing permissions and # limitations under the License. """Contains the noise perturb augmentation model.""" -from deepspeech.frontend.audio import AudioSegment -from deepspeech.frontend.augmentor.base import AugmentorBase -from deepspeech.frontend.utility import read_manifest +from paddlespeech.s2t.frontend.audio import AudioSegment +from paddlespeech.s2t.frontend.augmentor.base import AugmentorBase +from paddlespeech.s2t.frontend.utility import read_manifest class NoisePerturbAugmentor(AugmentorBase): diff --git a/deepspeech/frontend/augmentor/online_bayesian_normalization.py b/paddlespeech/s2t/frontend/augmentor/online_bayesian_normalization.py similarity index 97% rename from deepspeech/frontend/augmentor/online_bayesian_normalization.py rename to paddlespeech/s2t/frontend/augmentor/online_bayesian_normalization.py index 0f9d3ef6..f9d1530f 100644 --- a/deepspeech/frontend/augmentor/online_bayesian_normalization.py +++ b/paddlespeech/s2t/frontend/augmentor/online_bayesian_normalization.py @@ -12,7 +12,7 @@ # See the License for the specific language governing permissions and # limitations under the License. """Contain the online bayesian normalization augmentation model.""" -from deepspeech.frontend.augmentor.base import AugmentorBase +from paddlespeech.s2t.frontend.augmentor.base import AugmentorBase class OnlineBayesianNormalizationAugmentor(AugmentorBase): diff --git a/deepspeech/frontend/augmentor/resample.py b/paddlespeech/s2t/frontend/augmentor/resample.py similarity index 95% rename from deepspeech/frontend/augmentor/resample.py rename to paddlespeech/s2t/frontend/augmentor/resample.py index 509fe003..4e6402ff 100644 --- a/deepspeech/frontend/augmentor/resample.py +++ b/paddlespeech/s2t/frontend/augmentor/resample.py @@ -12,7 +12,7 @@ # See the License for the specific language governing permissions and # limitations under the License. """Contain the resample augmentation model.""" -from deepspeech.frontend.augmentor.base import AugmentorBase +from paddlespeech.s2t.frontend.augmentor.base import AugmentorBase class ResampleAugmentor(AugmentorBase): diff --git a/deepspeech/frontend/augmentor/shift_perturb.py b/paddlespeech/s2t/frontend/augmentor/shift_perturb.py similarity index 96% rename from deepspeech/frontend/augmentor/shift_perturb.py rename to paddlespeech/s2t/frontend/augmentor/shift_perturb.py index 8b7439fe..ed6f162b 100644 --- a/deepspeech/frontend/augmentor/shift_perturb.py +++ b/paddlespeech/s2t/frontend/augmentor/shift_perturb.py @@ -12,7 +12,7 @@ # See the License for the specific language governing permissions and # limitations under the License. """Contains the volume perturb augmentation model.""" -from deepspeech.frontend.augmentor.base import AugmentorBase +from paddlespeech.s2t.frontend.augmentor.base import AugmentorBase class ShiftPerturbAugmentor(AugmentorBase): diff --git a/deepspeech/frontend/augmentor/spec_augment.py b/paddlespeech/s2t/frontend/augmentor/spec_augment.py similarity index 98% rename from deepspeech/frontend/augmentor/spec_augment.py rename to paddlespeech/s2t/frontend/augmentor/spec_augment.py index e78f6f6a..e91cfdce 100644 --- a/deepspeech/frontend/augmentor/spec_augment.py +++ b/paddlespeech/s2t/frontend/augmentor/spec_augment.py @@ -18,8 +18,8 @@ import numpy as np from PIL import Image from PIL.Image import BICUBIC -from deepspeech.frontend.augmentor.base import AugmentorBase -from deepspeech.utils.log import Log +from paddlespeech.s2t.frontend.augmentor.base import AugmentorBase +from paddlespeech.s2t.utils.log import Log logger = Log(__name__).getlog() diff --git a/deepspeech/frontend/augmentor/speed_perturb.py b/paddlespeech/s2t/frontend/augmentor/speed_perturb.py similarity index 98% rename from deepspeech/frontend/augmentor/speed_perturb.py rename to paddlespeech/s2t/frontend/augmentor/speed_perturb.py index ce8dfde0..af0b23ee 100644 --- a/deepspeech/frontend/augmentor/speed_perturb.py +++ b/paddlespeech/s2t/frontend/augmentor/speed_perturb.py @@ -14,7 +14,7 @@ """Contain the speech perturbation augmentation model.""" import numpy as np -from deepspeech.frontend.augmentor.base import AugmentorBase +from paddlespeech.s2t.frontend.augmentor.base import AugmentorBase class SpeedPerturbAugmentor(AugmentorBase): diff --git a/deepspeech/frontend/augmentor/volume_perturb.py b/paddlespeech/s2t/frontend/augmentor/volume_perturb.py similarity index 96% rename from deepspeech/frontend/augmentor/volume_perturb.py rename to paddlespeech/s2t/frontend/augmentor/volume_perturb.py index 70cb2889..8cd2dc0a 100644 --- a/deepspeech/frontend/augmentor/volume_perturb.py +++ b/paddlespeech/s2t/frontend/augmentor/volume_perturb.py @@ -12,7 +12,7 @@ # See the License for the specific language governing permissions and # limitations under the License. """Contains the volume perturb augmentation model.""" -from deepspeech.frontend.augmentor.base import AugmentorBase +from paddlespeech.s2t.frontend.augmentor.base import AugmentorBase class VolumePerturbAugmentor(AugmentorBase): diff --git a/deepspeech/frontend/featurizer/__init__.py b/paddlespeech/s2t/frontend/featurizer/__init__.py similarity index 100% rename from deepspeech/frontend/featurizer/__init__.py rename to paddlespeech/s2t/frontend/featurizer/__init__.py diff --git a/deepspeech/frontend/featurizer/audio_featurizer.py b/paddlespeech/s2t/frontend/featurizer/audio_featurizer.py similarity index 100% rename from deepspeech/frontend/featurizer/audio_featurizer.py rename to paddlespeech/s2t/frontend/featurizer/audio_featurizer.py diff --git a/deepspeech/frontend/featurizer/speech_featurizer.py b/paddlespeech/s2t/frontend/featurizer/speech_featurizer.py similarity index 95% rename from deepspeech/frontend/featurizer/speech_featurizer.py rename to paddlespeech/s2t/frontend/featurizer/speech_featurizer.py index 25687140..591df96e 100644 --- a/deepspeech/frontend/featurizer/speech_featurizer.py +++ b/paddlespeech/s2t/frontend/featurizer/speech_featurizer.py @@ -12,8 +12,8 @@ # See the License for the specific language governing permissions and # limitations under the License. """Contains the speech featurizer class.""" -from deepspeech.frontend.featurizer.audio_featurizer import AudioFeaturizer -from deepspeech.frontend.featurizer.text_featurizer import TextFeaturizer +from paddlespeech.s2t.frontend.featurizer.audio_featurizer import AudioFeaturizer +from paddlespeech.s2t.frontend.featurizer.text_featurizer import TextFeaturizer class SpeechFeaturizer(): diff --git a/deepspeech/frontend/featurizer/text_featurizer.py b/paddlespeech/s2t/frontend/featurizer/text_featurizer.py similarity index 99% rename from deepspeech/frontend/featurizer/text_featurizer.py rename to paddlespeech/s2t/frontend/featurizer/text_featurizer.py index c596bd43..7f3bd9e1 100644 --- a/deepspeech/frontend/featurizer/text_featurizer.py +++ b/paddlespeech/s2t/frontend/featurizer/text_featurizer.py @@ -23,7 +23,7 @@ from ..utility import MASKCTC from ..utility import SOS from ..utility import SPACE from ..utility import UNK -from deepspeech.utils.log import Log +from paddlespeech.s2t.utils.log import Log logger = Log(__name__).getlog() diff --git a/deepspeech/frontend/normalizer.py b/paddlespeech/s2t/frontend/normalizer.py similarity index 96% rename from deepspeech/frontend/normalizer.py rename to paddlespeech/s2t/frontend/normalizer.py index 6ace4fc6..a29cddc3 100644 --- a/deepspeech/frontend/normalizer.py +++ b/paddlespeech/s2t/frontend/normalizer.py @@ -19,10 +19,10 @@ import paddle from paddle.io import DataLoader from paddle.io import Dataset -from deepspeech.frontend.audio import AudioSegment -from deepspeech.frontend.utility import load_cmvn -from deepspeech.frontend.utility import read_manifest -from deepspeech.utils.log import Log +from paddlespeech.s2t.frontend.audio import AudioSegment +from paddlespeech.s2t.frontend.utility import load_cmvn +from paddlespeech.s2t.frontend.utility import read_manifest +from paddlespeech.s2t.utils.log import Log __all__ = ["FeatureNormalizer"] diff --git a/deepspeech/frontend/speech.py b/paddlespeech/s2t/frontend/speech.py similarity index 99% rename from deepspeech/frontend/speech.py rename to paddlespeech/s2t/frontend/speech.py index 9eed9725..8fd661c9 100644 --- a/deepspeech/frontend/speech.py +++ b/paddlespeech/s2t/frontend/speech.py @@ -14,7 +14,7 @@ """Contains the speech segment class.""" import numpy as np -from deepspeech.frontend.audio import AudioSegment +from paddlespeech.s2t.frontend.audio import AudioSegment class SpeechSegment(AudioSegment): diff --git a/deepspeech/frontend/utility.py b/paddlespeech/s2t/frontend/utility.py similarity index 99% rename from deepspeech/frontend/utility.py rename to paddlespeech/s2t/frontend/utility.py index f83f1d4e..089890d2 100644 --- a/deepspeech/frontend/utility.py +++ b/paddlespeech/s2t/frontend/utility.py @@ -23,7 +23,7 @@ from typing import Text import jsonlines import numpy as np -from deepspeech.utils.log import Log +from paddlespeech.s2t.utils.log import Log logger = Log(__name__).getlog() diff --git a/deepspeech/io/__init__.py b/paddlespeech/s2t/io/__init__.py similarity index 100% rename from deepspeech/io/__init__.py rename to paddlespeech/s2t/io/__init__.py diff --git a/deepspeech/io/batchfy.py b/paddlespeech/s2t/io/batchfy.py similarity index 99% rename from deepspeech/io/batchfy.py rename to paddlespeech/s2t/io/batchfy.py index 06cb3c9d..3517f059 100644 --- a/deepspeech/io/batchfy.py +++ b/paddlespeech/s2t/io/batchfy.py @@ -15,7 +15,7 @@ import itertools import numpy as np -from deepspeech.utils.log import Log +from paddlespeech.s2t.utils.log import Log __all__ = ["make_batchset"] diff --git a/deepspeech/io/collator.py b/paddlespeech/s2t/io/collator.py similarity index 96% rename from deepspeech/io/collator.py rename to paddlespeech/s2t/io/collator.py index 5391260e..a500f10c 100644 --- a/deepspeech/io/collator.py +++ b/paddlespeech/s2t/io/collator.py @@ -17,16 +17,16 @@ from typing import Optional import numpy as np from yacs.config import CfgNode -from deepspeech.frontend.augmentor.augmentation import AugmentationPipeline -from deepspeech.frontend.featurizer.speech_featurizer import SpeechFeaturizer -from deepspeech.frontend.featurizer.text_featurizer import TextFeaturizer -from deepspeech.frontend.normalizer import FeatureNormalizer -from deepspeech.frontend.speech import SpeechSegment -from deepspeech.frontend.utility import IGNORE_ID -from deepspeech.frontend.utility import TarLocalData -from deepspeech.io.reader import LoadInputsAndTargets -from deepspeech.io.utility import pad_list -from deepspeech.utils.log import Log +from paddlespeech.s2t.frontend.augmentor.augmentation import AugmentationPipeline +from paddlespeech.s2t.frontend.featurizer.speech_featurizer import SpeechFeaturizer +from paddlespeech.s2t.frontend.featurizer.text_featurizer import TextFeaturizer +from paddlespeech.s2t.frontend.normalizer import FeatureNormalizer +from paddlespeech.s2t.frontend.speech import SpeechSegment +from paddlespeech.s2t.frontend.utility import IGNORE_ID +from paddlespeech.s2t.frontend.utility import TarLocalData +from paddlespeech.s2t.io.reader import LoadInputsAndTargets +from paddlespeech.s2t.io.utility import pad_list +from paddlespeech.s2t.utils.log import Log __all__ = ["SpeechCollator", "TripletSpeechCollator"] diff --git a/deepspeech/io/converter.py b/paddlespeech/s2t/io/converter.py similarity index 96% rename from deepspeech/io/converter.py rename to paddlespeech/s2t/io/converter.py index b80c7b20..94634360 100644 --- a/deepspeech/io/converter.py +++ b/paddlespeech/s2t/io/converter.py @@ -13,8 +13,8 @@ # limitations under the License. import numpy as np -from deepspeech.io.utility import pad_list -from deepspeech.utils.log import Log +from paddlespeech.s2t.io.utility import pad_list +from paddlespeech.s2t.utils.log import Log __all__ = ["CustomConverter"] diff --git a/deepspeech/io/dataloader.py b/paddlespeech/s2t/io/dataloader.py similarity index 94% rename from deepspeech/io/dataloader.py rename to paddlespeech/s2t/io/dataloader.py index d8ef9ba6..3b5000a2 100644 --- a/deepspeech/io/dataloader.py +++ b/paddlespeech/s2t/io/dataloader.py @@ -19,12 +19,12 @@ from typing import Text import numpy as np from paddle.io import DataLoader -from deepspeech.frontend.utility import read_manifest -from deepspeech.io.batchfy import make_batchset -from deepspeech.io.converter import CustomConverter -from deepspeech.io.dataset import TransformDataset -from deepspeech.io.reader import LoadInputsAndTargets -from deepspeech.utils.log import Log +from paddlespeech.s2t.frontend.utility import read_manifest +from paddlespeech.s2t.io.batchfy import make_batchset +from paddlespeech.s2t.io.converter import CustomConverter +from paddlespeech.s2t.io.dataset import TransformDataset +from paddlespeech.s2t.io.reader import LoadInputsAndTargets +from paddlespeech.s2t.utils.log import Log __all__ = ["BatchDataLoader"] diff --git a/deepspeech/io/dataset.py b/paddlespeech/s2t/io/dataset.py similarity index 99% rename from deepspeech/io/dataset.py rename to paddlespeech/s2t/io/dataset.py index 7c101002..121410c8 100644 --- a/deepspeech/io/dataset.py +++ b/paddlespeech/s2t/io/dataset.py @@ -16,8 +16,8 @@ from typing import Optional from paddle.io import Dataset from yacs.config import CfgNode -from deepspeech.frontend.utility import read_manifest -from deepspeech.utils.log import Log +from paddlespeech.s2t.frontend.utility import read_manifest +from paddlespeech.s2t.utils.log import Log __all__ = ["ManifestDataset", "TransformDataset"] diff --git a/deepspeech/io/reader.py b/paddlespeech/s2t/io/reader.py similarity index 99% rename from deepspeech/io/reader.py rename to paddlespeech/s2t/io/reader.py index 59098752..faf2f600 100644 --- a/deepspeech/io/reader.py +++ b/paddlespeech/s2t/io/reader.py @@ -17,8 +17,8 @@ import kaldiio import numpy as np import soundfile -from deepspeech.frontend.augmentor.augmentation import AugmentationPipeline as Transformation -from deepspeech.utils.log import Log +from paddlespeech.s2t.frontend.augmentor.augmentation import AugmentationPipeline as Transformation +from paddlespeech.s2t.utils.log import Log __all__ = ["LoadInputsAndTargets"] diff --git a/deepspeech/io/sampler.py b/paddlespeech/s2t/io/sampler.py similarity index 99% rename from deepspeech/io/sampler.py rename to paddlespeech/s2t/io/sampler.py index 763a3781..35b57524 100644 --- a/deepspeech/io/sampler.py +++ b/paddlespeech/s2t/io/sampler.py @@ -18,7 +18,7 @@ from paddle import distributed as dist from paddle.io import BatchSampler from paddle.io import DistributedBatchSampler -from deepspeech.utils.log import Log +from paddlespeech.s2t.utils.log import Log __all__ = [ "SortagradDistributedBatchSampler", diff --git a/deepspeech/io/utility.py b/paddlespeech/s2t/io/utility.py similarity index 98% rename from deepspeech/io/utility.py rename to paddlespeech/s2t/io/utility.py index 99487a0a..392031ba 100644 --- a/deepspeech/io/utility.py +++ b/paddlespeech/s2t/io/utility.py @@ -15,7 +15,7 @@ from typing import List import numpy as np -from deepspeech.utils.log import Log +from paddlespeech.s2t.utils.log import Log __all__ = ["pad_list", "pad_sequence"] diff --git a/deepspeech/models/__init__.py b/paddlespeech/s2t/models/__init__.py similarity index 100% rename from deepspeech/models/__init__.py rename to paddlespeech/s2t/models/__init__.py diff --git a/deepspeech/models/asr_interface.py b/paddlespeech/s2t/models/asr_interface.py similarity index 94% rename from deepspeech/models/asr_interface.py rename to paddlespeech/s2t/models/asr_interface.py index d86daa0b..ece83e51 100644 --- a/deepspeech/models/asr_interface.py +++ b/paddlespeech/s2t/models/asr_interface.py @@ -14,7 +14,7 @@ """ASR Interface module.""" import argparse -from deepspeech.utils.dynamic_import import dynamic_import +from paddlespeech.s2t.utils.dynamic_import import dynamic_import class ASRInterface: @@ -103,14 +103,14 @@ class ASRInterface: @property def attention_plot_class(self): """Get attention plot class.""" - from deepspeech.training.extensions.plot import PlotAttentionReport + from paddlespeech.s2t.training.extensions.plot import PlotAttentionReport return PlotAttentionReport @property def ctc_plot_class(self): """Get CTC plot class.""" - from deepspeech.training.extensions.plot import PlotCTCReport + from paddlespeech.s2t.training.extensions.plot import PlotCTCReport return PlotCTCReport @@ -140,8 +140,8 @@ class ASRInterface: predefined_asr = { - "transformer": "deepspeech.models.u2:U2Model", - "conformer": "deepspeech.models.u2:U2Model", + "transformer": "paddlespeech.s2t.models.u2:U2Model", + "conformer": "paddlespeech.s2t.models.u2:U2Model", } diff --git a/deepspeech/models/ds2/__init__.py b/paddlespeech/s2t/models/ds2/__init__.py similarity index 100% rename from deepspeech/models/ds2/__init__.py rename to paddlespeech/s2t/models/ds2/__init__.py diff --git a/deepspeech/models/ds2/conv.py b/paddlespeech/s2t/models/ds2/conv.py similarity index 97% rename from deepspeech/models/ds2/conv.py rename to paddlespeech/s2t/models/ds2/conv.py index 069b7dd4..4e766e79 100644 --- a/deepspeech/models/ds2/conv.py +++ b/paddlespeech/s2t/models/ds2/conv.py @@ -14,9 +14,9 @@ from paddle import nn from paddle.nn import functional as F -from deepspeech.modules.activation import brelu -from deepspeech.modules.mask import make_non_pad_mask -from deepspeech.utils.log import Log +from paddlespeech.s2t.modules.activation import brelu +from paddlespeech.s2t.modules.mask import make_non_pad_mask +from paddlespeech.s2t.utils.log import Log logger = Log(__name__).getlog() diff --git a/deepspeech/models/ds2/deepspeech2.py b/paddlespeech/s2t/models/ds2/deepspeech2.py similarity index 97% rename from deepspeech/models/ds2/deepspeech2.py rename to paddlespeech/s2t/models/ds2/deepspeech2.py index 63327a8c..4a7a7c15 100644 --- a/deepspeech/models/ds2/deepspeech2.py +++ b/paddlespeech/s2t/models/ds2/deepspeech2.py @@ -18,12 +18,12 @@ import paddle from paddle import nn from yacs.config import CfgNode -from deepspeech.models.ds2.conv import ConvStack -from deepspeech.models.ds2.rnn import RNNStack -from deepspeech.modules.ctc import CTCDecoder -from deepspeech.utils import layer_tools -from deepspeech.utils.checkpoint import Checkpoint -from deepspeech.utils.log import Log +from paddlespeech.s2t.models.ds2.conv import ConvStack +from paddlespeech.s2t.models.ds2.rnn import RNNStack +from paddlespeech.s2t.modules.ctc import CTCDecoder +from paddlespeech.s2t.utils import layer_tools +from paddlespeech.s2t.utils.checkpoint import Checkpoint +from paddlespeech.s2t.utils.log import Log logger = Log(__name__).getlog() diff --git a/deepspeech/models/ds2/rnn.py b/paddlespeech/s2t/models/ds2/rnn.py similarity index 98% rename from deepspeech/models/ds2/rnn.py rename to paddlespeech/s2t/models/ds2/rnn.py index d6a8b394..f655b2d8 100644 --- a/deepspeech/models/ds2/rnn.py +++ b/paddlespeech/s2t/models/ds2/rnn.py @@ -18,9 +18,9 @@ from paddle import nn from paddle.nn import functional as F from paddle.nn import initializer as I -from deepspeech.modules.activation import brelu -from deepspeech.modules.mask import make_non_pad_mask -from deepspeech.utils.log import Log +from paddlespeech.s2t.modules.activation import brelu +from paddlespeech.s2t.modules.mask import make_non_pad_mask +from paddlespeech.s2t.utils.log import Log logger = Log(__name__).getlog() diff --git a/deepspeech/models/ds2_online/__init__.py b/paddlespeech/s2t/models/ds2_online/__init__.py similarity index 100% rename from deepspeech/models/ds2_online/__init__.py rename to paddlespeech/s2t/models/ds2_online/__init__.py diff --git a/deepspeech/models/ds2_online/conv.py b/paddlespeech/s2t/models/ds2_online/conv.py similarity index 95% rename from deepspeech/models/ds2_online/conv.py rename to paddlespeech/s2t/models/ds2_online/conv.py index 4a6fd5ab..25a9715a 100644 --- a/deepspeech/models/ds2_online/conv.py +++ b/paddlespeech/s2t/models/ds2_online/conv.py @@ -13,7 +13,7 @@ # limitations under the License. import paddle -from deepspeech.modules.subsampling import Conv2dSubsampling4 +from paddlespeech.s2t.modules.subsampling import Conv2dSubsampling4 class Conv2dSubsampling4Online(Conv2dSubsampling4): diff --git a/deepspeech/models/ds2_online/deepspeech2.py b/paddlespeech/s2t/models/ds2_online/deepspeech2.py similarity index 98% rename from deepspeech/models/ds2_online/deepspeech2.py rename to paddlespeech/s2t/models/ds2_online/deepspeech2.py index 52e0c7b1..da04d5c5 100644 --- a/deepspeech/models/ds2_online/deepspeech2.py +++ b/paddlespeech/s2t/models/ds2_online/deepspeech2.py @@ -19,11 +19,11 @@ import paddle.nn.functional as F from paddle import nn from yacs.config import CfgNode -from deepspeech.models.ds2_online.conv import Conv2dSubsampling4Online -from deepspeech.modules.ctc import CTCDecoder -from deepspeech.utils import layer_tools -from deepspeech.utils.checkpoint import Checkpoint -from deepspeech.utils.log import Log +from paddlespeech.s2t.models.ds2_online.conv import Conv2dSubsampling4Online +from paddlespeech.s2t.modules.ctc import CTCDecoder +from paddlespeech.s2t.utils import layer_tools +from paddlespeech.s2t.utils.checkpoint import Checkpoint +from paddlespeech.s2t.utils.log import Log logger = Log(__name__).getlog() __all__ = ['DeepSpeech2ModelOnline', 'DeepSpeech2InferModelOnline'] diff --git a/deepspeech/models/lm/__init__.py b/paddlespeech/s2t/models/lm/__init__.py similarity index 100% rename from deepspeech/models/lm/__init__.py rename to paddlespeech/s2t/models/lm/__init__.py diff --git a/deepspeech/models/lm/transformer.py b/paddlespeech/s2t/models/lm/transformer.py similarity index 96% rename from deepspeech/models/lm/transformer.py rename to paddlespeech/s2t/models/lm/transformer.py index 19e2b758..447f1dbc 100644 --- a/deepspeech/models/lm/transformer.py +++ b/paddlespeech/s2t/models/lm/transformer.py @@ -20,11 +20,11 @@ import paddle import paddle.nn as nn import paddle.nn.functional as F -from deepspeech.decoders.scorers.scorer_interface import BatchScorerInterface -from deepspeech.models.lm_interface import LMInterface -from deepspeech.modules.encoder import TransformerEncoder -from deepspeech.modules.mask import subsequent_mask -from deepspeech.utils.log import Log +from paddlespeech.s2t.decoders.scorers.scorer_interface import BatchScorerInterface +from paddlespeech.s2t.models.lm_interface import LMInterface +from paddlespeech.s2t.modules.encoder import TransformerEncoder +from paddlespeech.s2t.modules.mask import subsequent_mask +from paddlespeech.s2t.utils.log import Log logger = Log(__name__).getlog() diff --git a/deepspeech/models/lm_interface.py b/paddlespeech/s2t/models/lm_interface.py similarity index 91% rename from deepspeech/models/lm_interface.py rename to paddlespeech/s2t/models/lm_interface.py index e2987282..7ae39b35 100644 --- a/deepspeech/models/lm_interface.py +++ b/paddlespeech/s2t/models/lm_interface.py @@ -14,8 +14,8 @@ """Language model interface.""" import argparse -from deepspeech.decoders.scorers.scorer_interface import ScorerInterface -from deepspeech.utils.dynamic_import import dynamic_import +from paddlespeech.s2t.decoders.scorers.scorer_interface import ScorerInterface +from paddlespeech.s2t.utils.dynamic_import import dynamic_import class LMInterface(ScorerInterface): @@ -62,7 +62,7 @@ class LMInterface(ScorerInterface): predefined_lms = { - "transformer": "deepspeech.models.lm.transformer:TransformerLM", + "transformer": "paddlespeech.s2t.models.lm.transformer:TransformerLM", } diff --git a/deepspeech/models/st_interface.py b/paddlespeech/s2t/models/st_interface.py similarity index 95% rename from deepspeech/models/st_interface.py rename to paddlespeech/s2t/models/st_interface.py index 05939f9a..331f1e82 100644 --- a/deepspeech/models/st_interface.py +++ b/paddlespeech/s2t/models/st_interface.py @@ -13,7 +13,7 @@ # limitations under the License. """ST Interface module.""" from .asr_interface import ASRInterface -from deepspeech.utils.dynamic_import import dynamic_import +from paddlespeech.s2t.utils.dynamic_import import dynamic_import class STInterface(ASRInterface): @@ -55,7 +55,7 @@ class STInterface(ASRInterface): predefined_st = { - "transformer": "deepspeech.models.u2_st:U2STModel", + "transformer": "paddlespeech.s2t.models.u2_st:U2STModel", } diff --git a/deepspeech/models/u2/__init__.py b/paddlespeech/s2t/models/u2/__init__.py similarity index 100% rename from deepspeech/models/u2/__init__.py rename to paddlespeech/s2t/models/u2/__init__.py diff --git a/deepspeech/models/u2/u2.py b/paddlespeech/s2t/models/u2/u2.py similarity index 96% rename from deepspeech/models/u2/u2.py rename to paddlespeech/s2t/models/u2/u2.py index 6cd3b775..fd998271 100644 --- a/deepspeech/models/u2/u2.py +++ b/paddlespeech/s2t/models/u2/u2.py @@ -28,29 +28,29 @@ from paddle import jit from paddle import nn from yacs.config import CfgNode -from deepspeech.decoders.scorers.ctc import CTCPrefixScorer -from deepspeech.frontend.utility import IGNORE_ID -from deepspeech.frontend.utility import load_cmvn -from deepspeech.models.asr_interface import ASRInterface -from deepspeech.modules.cmvn import GlobalCMVN -from deepspeech.modules.ctc import CTCDecoder -from deepspeech.modules.decoder import TransformerDecoder -from deepspeech.modules.encoder import ConformerEncoder -from deepspeech.modules.encoder import TransformerEncoder -from deepspeech.modules.loss import LabelSmoothingLoss -from deepspeech.modules.mask import make_pad_mask -from deepspeech.modules.mask import mask_finished_preds -from deepspeech.modules.mask import mask_finished_scores -from deepspeech.modules.mask import subsequent_mask -from deepspeech.utils import checkpoint -from deepspeech.utils import layer_tools -from deepspeech.utils.ctc_utils import remove_duplicates_and_blank -from deepspeech.utils.log import Log -from deepspeech.utils.tensor_utils import add_sos_eos -from deepspeech.utils.tensor_utils import pad_sequence -from deepspeech.utils.tensor_utils import th_accuracy -from deepspeech.utils.utility import log_add -from deepspeech.utils.utility import UpdateConfig +from paddlespeech.s2t.decoders.scorers.ctc import CTCPrefixScorer +from paddlespeech.s2t.frontend.utility import IGNORE_ID +from paddlespeech.s2t.frontend.utility import load_cmvn +from paddlespeech.s2t.models.asr_interface import ASRInterface +from paddlespeech.s2t.modules.cmvn import GlobalCMVN +from paddlespeech.s2t.modules.ctc import CTCDecoder +from paddlespeech.s2t.modules.decoder import TransformerDecoder +from paddlespeech.s2t.modules.encoder import ConformerEncoder +from paddlespeech.s2t.modules.encoder import TransformerEncoder +from paddlespeech.s2t.modules.loss import LabelSmoothingLoss +from paddlespeech.s2t.modules.mask import make_pad_mask +from paddlespeech.s2t.modules.mask import mask_finished_preds +from paddlespeech.s2t.modules.mask import mask_finished_scores +from paddlespeech.s2t.modules.mask import subsequent_mask +from paddlespeech.s2t.utils import checkpoint +from paddlespeech.s2t.utils import layer_tools +from paddlespeech.s2t.utils.ctc_utils import remove_duplicates_and_blank +from paddlespeech.s2t.utils.log import Log +from paddlespeech.s2t.utils.tensor_utils import add_sos_eos +from paddlespeech.s2t.utils.tensor_utils import pad_sequence +from paddlespeech.s2t.utils.tensor_utils import th_accuracy +from paddlespeech.s2t.utils.utility import log_add +from paddlespeech.s2t.utils.utility import UpdateConfig __all__ = ["U2Model", "U2InferModel"] diff --git a/deepspeech/models/u2/updater.py b/paddlespeech/s2t/models/u2/updater.py similarity index 93% rename from deepspeech/models/u2/updater.py rename to paddlespeech/s2t/models/u2/updater.py index 7b70ca04..bb18fe41 100644 --- a/deepspeech/models/u2/updater.py +++ b/paddlespeech/s2t/models/u2/updater.py @@ -16,12 +16,12 @@ from contextlib import nullcontext import paddle from paddle import distributed as dist -from deepspeech.training.extensions.evaluator import StandardEvaluator -from deepspeech.training.reporter import report -from deepspeech.training.timer import Timer -from deepspeech.training.updaters.standard_updater import StandardUpdater -from deepspeech.utils import layer_tools -from deepspeech.utils.log import Log +from paddlespeech.s2t.training.extensions.evaluator import StandardEvaluator +from paddlespeech.s2t.training.reporter import report +from paddlespeech.s2t.training.timer import Timer +from paddlespeech.s2t.training.updaters.standard_updater import StandardUpdater +from paddlespeech.s2t.utils import layer_tools +from paddlespeech.s2t.utils.log import Log logger = Log(__name__).getlog() diff --git a/deepspeech/models/u2_st/__init__.py b/paddlespeech/s2t/models/u2_st/__init__.py similarity index 100% rename from deepspeech/models/u2_st/__init__.py rename to paddlespeech/s2t/models/u2_st/__init__.py diff --git a/deepspeech/models/u2_st/u2_st.py b/paddlespeech/s2t/models/u2_st/u2_st.py similarity index 96% rename from deepspeech/models/u2_st/u2_st.py rename to paddlespeech/s2t/models/u2_st/u2_st.py index bf98423d..76c50150 100644 --- a/deepspeech/models/u2_st/u2_st.py +++ b/paddlespeech/s2t/models/u2_st/u2_st.py @@ -26,23 +26,23 @@ from paddle import jit from paddle import nn from yacs.config import CfgNode -from deepspeech.frontend.utility import IGNORE_ID -from deepspeech.frontend.utility import load_cmvn -from deepspeech.modules.cmvn import GlobalCMVN -from deepspeech.modules.ctc import CTCDecoder -from deepspeech.modules.decoder import TransformerDecoder -from deepspeech.modules.encoder import ConformerEncoder -from deepspeech.modules.encoder import TransformerEncoder -from deepspeech.modules.loss import LabelSmoothingLoss -from deepspeech.modules.mask import mask_finished_preds -from deepspeech.modules.mask import mask_finished_scores -from deepspeech.modules.mask import subsequent_mask -from deepspeech.utils import checkpoint -from deepspeech.utils import layer_tools -from deepspeech.utils.log import Log -from deepspeech.utils.tensor_utils import add_sos_eos -from deepspeech.utils.tensor_utils import th_accuracy -from deepspeech.utils.utility import UpdateConfig +from paddlespeech.s2t.frontend.utility import IGNORE_ID +from paddlespeech.s2t.frontend.utility import load_cmvn +from paddlespeech.s2t.modules.cmvn import GlobalCMVN +from paddlespeech.s2t.modules.ctc import CTCDecoder +from paddlespeech.s2t.modules.decoder import TransformerDecoder +from paddlespeech.s2t.modules.encoder import ConformerEncoder +from paddlespeech.s2t.modules.encoder import TransformerEncoder +from paddlespeech.s2t.modules.loss import LabelSmoothingLoss +from paddlespeech.s2t.modules.mask import mask_finished_preds +from paddlespeech.s2t.modules.mask import mask_finished_scores +from paddlespeech.s2t.modules.mask import subsequent_mask +from paddlespeech.s2t.utils import checkpoint +from paddlespeech.s2t.utils import layer_tools +from paddlespeech.s2t.utils.log import Log +from paddlespeech.s2t.utils.tensor_utils import add_sos_eos +from paddlespeech.s2t.utils.tensor_utils import th_accuracy +from paddlespeech.s2t.utils.utility import UpdateConfig __all__ = ["U2STModel", "U2STInferModel"] diff --git a/deepspeech/modules/__init__.py b/paddlespeech/s2t/modules/__init__.py similarity index 100% rename from deepspeech/modules/__init__.py rename to paddlespeech/s2t/modules/__init__.py diff --git a/deepspeech/modules/activation.py b/paddlespeech/s2t/modules/activation.py similarity index 99% rename from deepspeech/modules/activation.py rename to paddlespeech/s2t/modules/activation.py index 3cb8729e..4081f7f8 100644 --- a/deepspeech/modules/activation.py +++ b/paddlespeech/s2t/modules/activation.py @@ -17,7 +17,7 @@ import paddle from paddle import nn from paddle.nn import functional as F -from deepspeech.utils.log import Log +from paddlespeech.s2t.utils.log import Log logger = Log(__name__).getlog() diff --git a/deepspeech/modules/attention.py b/paddlespeech/s2t/modules/attention.py similarity index 99% rename from deepspeech/modules/attention.py rename to paddlespeech/s2t/modules/attention.py index f9479728..d5246777 100644 --- a/deepspeech/modules/attention.py +++ b/paddlespeech/s2t/modules/attention.py @@ -20,7 +20,7 @@ import paddle from paddle import nn from paddle.nn import initializer as I -from deepspeech.utils.log import Log +from paddlespeech.s2t.utils.log import Log logger = Log(__name__).getlog() diff --git a/deepspeech/modules/cmvn.py b/paddlespeech/s2t/modules/cmvn.py similarity index 97% rename from deepspeech/modules/cmvn.py rename to paddlespeech/s2t/modules/cmvn.py index b98af052..9aa938af 100644 --- a/deepspeech/modules/cmvn.py +++ b/paddlespeech/s2t/modules/cmvn.py @@ -14,7 +14,7 @@ import paddle from paddle import nn -from deepspeech.utils.log import Log +from paddlespeech.s2t.utils.log import Log logger = Log(__name__).getlog() diff --git a/deepspeech/modules/conformer_convolution.py b/paddlespeech/s2t/modules/conformer_convolution.py similarity index 99% rename from deepspeech/modules/conformer_convolution.py rename to paddlespeech/s2t/modules/conformer_convolution.py index 59249d2a..39b1ee32 100644 --- a/deepspeech/modules/conformer_convolution.py +++ b/paddlespeech/s2t/modules/conformer_convolution.py @@ -19,7 +19,7 @@ import paddle from paddle import nn from typeguard import check_argument_types -from deepspeech.utils.log import Log +from paddlespeech.s2t.utils.log import Log logger = Log(__name__).getlog() diff --git a/deepspeech/modules/crf.py b/paddlespeech/s2t/modules/crf.py similarity index 99% rename from deepspeech/modules/crf.py rename to paddlespeech/s2t/modules/crf.py index b6b481a0..66f6b180 100644 --- a/deepspeech/modules/crf.py +++ b/paddlespeech/s2t/modules/crf.py @@ -14,7 +14,7 @@ import paddle from paddle import nn -from deepspeech.utils.log import Log +from paddlespeech.s2t.utils.log import Log logger = Log(__name__).getlog() diff --git a/deepspeech/modules/ctc.py b/paddlespeech/s2t/modules/ctc.py similarity index 96% rename from deepspeech/modules/ctc.py rename to paddlespeech/s2t/modules/ctc.py index df6848db..120abd2b 100644 --- a/deepspeech/modules/ctc.py +++ b/paddlespeech/s2t/modules/ctc.py @@ -18,16 +18,16 @@ from paddle import nn from paddle.nn import functional as F from typeguard import check_argument_types -from deepspeech.modules.loss import CTCLoss -from deepspeech.utils import ctc_utils -from deepspeech.utils.log import Log +from paddlespeech.s2t.modules.loss import CTCLoss +from paddlespeech.s2t.utils import ctc_utils +from paddlespeech.s2t.utils.log import Log logger = Log(__name__).getlog() try: - from deepspeech.decoders.ctcdecoder.swig_wrapper import ctc_beam_search_decoder_batch # noqa: F401 - from deepspeech.decoders.ctcdecoder.swig_wrapper import ctc_greedy_decoder # noqa: F401 - from deepspeech.decoders.ctcdecoder.swig_wrapper import Scorer # noqa: F401 + from paddlespeech.s2t.decoders.ctcdecoder.swig_wrapper import ctc_beam_search_decoder_batch # noqa: F401 + from paddlespeech.s2t.decoders.ctcdecoder.swig_wrapper import ctc_greedy_decoder # noqa: F401 + from paddlespeech.s2t.decoders.ctcdecoder.swig_wrapper import Scorer # noqa: F401 except Exception as e: logger.info("ctcdecoder not installed!") diff --git a/deepspeech/modules/decoder.py b/paddlespeech/s2t/modules/decoder.py similarity index 94% rename from deepspeech/modules/decoder.py rename to paddlespeech/s2t/modules/decoder.py index 735f06dc..7d545f5e 100644 --- a/deepspeech/modules/decoder.py +++ b/paddlespeech/s2t/modules/decoder.py @@ -21,15 +21,15 @@ import paddle from paddle import nn from typeguard import check_argument_types -from deepspeech.decoders.scorers.scorer_interface import BatchScorerInterface -from deepspeech.modules.attention import MultiHeadedAttention -from deepspeech.modules.decoder_layer import DecoderLayer -from deepspeech.modules.embedding import PositionalEncoding -from deepspeech.modules.mask import make_non_pad_mask -from deepspeech.modules.mask import make_xs_mask -from deepspeech.modules.mask import subsequent_mask -from deepspeech.modules.positionwise_feed_forward import PositionwiseFeedForward -from deepspeech.utils.log import Log +from paddlespeech.s2t.decoders.scorers.scorer_interface import BatchScorerInterface +from paddlespeech.s2t.modules.attention import MultiHeadedAttention +from paddlespeech.s2t.modules.decoder_layer import DecoderLayer +from paddlespeech.s2t.modules.embedding import PositionalEncoding +from paddlespeech.s2t.modules.mask import make_non_pad_mask +from paddlespeech.s2t.modules.mask import make_xs_mask +from paddlespeech.s2t.modules.mask import subsequent_mask +from paddlespeech.s2t.modules.positionwise_feed_forward import PositionwiseFeedForward +from paddlespeech.s2t.utils.log import Log logger = Log(__name__).getlog() diff --git a/deepspeech/modules/decoder_layer.py b/paddlespeech/s2t/modules/decoder_layer.py similarity index 99% rename from deepspeech/modules/decoder_layer.py rename to paddlespeech/s2t/modules/decoder_layer.py index 47c42615..816bc9f6 100644 --- a/deepspeech/modules/decoder_layer.py +++ b/paddlespeech/s2t/modules/decoder_layer.py @@ -18,7 +18,7 @@ from typing import Tuple import paddle from paddle import nn -from deepspeech.utils.log import Log +from paddlespeech.s2t.utils.log import Log logger = Log(__name__).getlog() diff --git a/deepspeech/modules/embedding.py b/paddlespeech/s2t/modules/embedding.py similarity index 99% rename from deepspeech/modules/embedding.py rename to paddlespeech/s2t/modules/embedding.py index 64d594c2..938898c1 100644 --- a/deepspeech/modules/embedding.py +++ b/paddlespeech/s2t/modules/embedding.py @@ -18,7 +18,7 @@ from typing import Tuple import paddle from paddle import nn -from deepspeech.utils.log import Log +from paddlespeech.s2t.utils.log import Log logger = Log(__name__).getlog() diff --git a/deepspeech/modules/encoder.py b/paddlespeech/s2t/modules/encoder.py similarity index 94% rename from deepspeech/modules/encoder.py rename to paddlespeech/s2t/modules/encoder.py index 435b6894..246536db 100644 --- a/deepspeech/modules/encoder.py +++ b/paddlespeech/s2t/modules/encoder.py @@ -20,23 +20,23 @@ import paddle from paddle import nn from typeguard import check_argument_types -from deepspeech.modules.activation import get_activation -from deepspeech.modules.attention import MultiHeadedAttention -from deepspeech.modules.attention import RelPositionMultiHeadedAttention -from deepspeech.modules.conformer_convolution import ConvolutionModule -from deepspeech.modules.embedding import NoPositionalEncoding -from deepspeech.modules.embedding import PositionalEncoding -from deepspeech.modules.embedding import RelPositionalEncoding -from deepspeech.modules.encoder_layer import ConformerEncoderLayer -from deepspeech.modules.encoder_layer import TransformerEncoderLayer -from deepspeech.modules.mask import add_optional_chunk_mask -from deepspeech.modules.mask import make_non_pad_mask -from deepspeech.modules.positionwise_feed_forward import PositionwiseFeedForward -from deepspeech.modules.subsampling import Conv2dSubsampling4 -from deepspeech.modules.subsampling import Conv2dSubsampling6 -from deepspeech.modules.subsampling import Conv2dSubsampling8 -from deepspeech.modules.subsampling import LinearNoSubsampling -from deepspeech.utils.log import Log +from paddlespeech.s2t.modules.activation import get_activation +from paddlespeech.s2t.modules.attention import MultiHeadedAttention +from paddlespeech.s2t.modules.attention import RelPositionMultiHeadedAttention +from paddlespeech.s2t.modules.conformer_convolution import ConvolutionModule +from paddlespeech.s2t.modules.embedding import NoPositionalEncoding +from paddlespeech.s2t.modules.embedding import PositionalEncoding +from paddlespeech.s2t.modules.embedding import RelPositionalEncoding +from paddlespeech.s2t.modules.encoder_layer import ConformerEncoderLayer +from paddlespeech.s2t.modules.encoder_layer import TransformerEncoderLayer +from paddlespeech.s2t.modules.mask import add_optional_chunk_mask +from paddlespeech.s2t.modules.mask import make_non_pad_mask +from paddlespeech.s2t.modules.positionwise_feed_forward import PositionwiseFeedForward +from paddlespeech.s2t.modules.subsampling import Conv2dSubsampling4 +from paddlespeech.s2t.modules.subsampling import Conv2dSubsampling6 +from paddlespeech.s2t.modules.subsampling import Conv2dSubsampling8 +from paddlespeech.s2t.modules.subsampling import LinearNoSubsampling +from paddlespeech.s2t.utils.log import Log logger = Log(__name__).getlog() diff --git a/deepspeech/modules/encoder_layer.py b/paddlespeech/s2t/modules/encoder_layer.py similarity index 99% rename from deepspeech/modules/encoder_layer.py rename to paddlespeech/s2t/modules/encoder_layer.py index 6f49cfc8..0525a614 100644 --- a/deepspeech/modules/encoder_layer.py +++ b/paddlespeech/s2t/modules/encoder_layer.py @@ -18,7 +18,7 @@ from typing import Tuple import paddle from paddle import nn -from deepspeech.utils.log import Log +from paddlespeech.s2t.utils.log import Log logger = Log(__name__).getlog() diff --git a/deepspeech/modules/loss.py b/paddlespeech/s2t/modules/loss.py similarity index 99% rename from deepspeech/modules/loss.py rename to paddlespeech/s2t/modules/loss.py index e1138810..012ef242 100644 --- a/deepspeech/modules/loss.py +++ b/paddlespeech/s2t/modules/loss.py @@ -17,7 +17,7 @@ import paddle from paddle import nn from paddle.nn import functional as F -from deepspeech.utils.log import Log +from paddlespeech.s2t.utils.log import Log logger = Log(__name__).getlog() diff --git a/deepspeech/modules/mask.py b/paddlespeech/s2t/modules/mask.py similarity index 99% rename from deepspeech/modules/mask.py rename to paddlespeech/s2t/modules/mask.py index 52f8e4bc..5c7886d4 100644 --- a/deepspeech/modules/mask.py +++ b/paddlespeech/s2t/modules/mask.py @@ -13,7 +13,7 @@ # limitations under the License. import paddle -from deepspeech.utils.log import Log +from paddlespeech.s2t.utils.log import Log logger = Log(__name__).getlog() diff --git a/deepspeech/modules/positionwise_feed_forward.py b/paddlespeech/s2t/modules/positionwise_feed_forward.py similarity index 97% rename from deepspeech/modules/positionwise_feed_forward.py rename to paddlespeech/s2t/modules/positionwise_feed_forward.py index 9732e1aa..ccea8e57 100644 --- a/deepspeech/modules/positionwise_feed_forward.py +++ b/paddlespeech/s2t/modules/positionwise_feed_forward.py @@ -15,7 +15,7 @@ import paddle from paddle import nn -from deepspeech.utils.log import Log +from paddlespeech.s2t.utils.log import Log logger = Log(__name__).getlog() diff --git a/deepspeech/modules/subsampling.py b/paddlespeech/s2t/modules/subsampling.py similarity index 98% rename from deepspeech/modules/subsampling.py rename to paddlespeech/s2t/modules/subsampling.py index 13e2c8ef..6b8c28c6 100644 --- a/deepspeech/modules/subsampling.py +++ b/paddlespeech/s2t/modules/subsampling.py @@ -17,8 +17,8 @@ from typing import Tuple import paddle from paddle import nn -from deepspeech.modules.embedding import PositionalEncoding -from deepspeech.utils.log import Log +from paddlespeech.s2t.modules.embedding import PositionalEncoding +from paddlespeech.s2t.utils.log import Log logger = Log(__name__).getlog() diff --git a/deepspeech/training/__init__.py b/paddlespeech/s2t/training/__init__.py similarity index 100% rename from deepspeech/training/__init__.py rename to paddlespeech/s2t/training/__init__.py diff --git a/deepspeech/training/cli.py b/paddlespeech/s2t/training/cli.py similarity index 95% rename from deepspeech/training/cli.py rename to paddlespeech/s2t/training/cli.py index 14a34cb7..55b010e9 100644 --- a/deepspeech/training/cli.py +++ b/paddlespeech/s2t/training/cli.py @@ -36,10 +36,10 @@ class LoadFromFile(argparse.Action): def default_argument_parser(parser=None): - r"""A simple yet genral argument parser for experiments with parakeet. + r"""A simple yet genral argument parser for experiments with t2s. - This is used in examples with parakeet. And it is intended to be used by - other experiments with parakeet. It requires a minimal set of command line + This is used in examples with t2s. And it is intended to be used by + other experiments with t2s. It requires a minimal set of command line arguments to start a training script. The ``--config`` and ``--opts`` are used for overwrite the deault @@ -56,7 +56,7 @@ def default_argument_parser(parser=None): See Also -------- - parakeet.training.experiment + paddlespeech.t2s.training.experiment Returns ------- argparse.ArgumentParser diff --git a/deepspeech/training/extensions/__init__.py b/paddlespeech/s2t/training/extensions/__init__.py similarity index 100% rename from deepspeech/training/extensions/__init__.py rename to paddlespeech/s2t/training/extensions/__init__.py diff --git a/deepspeech/training/extensions/evaluator.py b/paddlespeech/s2t/training/extensions/evaluator.py similarity index 98% rename from deepspeech/training/extensions/evaluator.py rename to paddlespeech/s2t/training/extensions/evaluator.py index 1026a4ec..9da0d731 100644 --- a/deepspeech/training/extensions/evaluator.py +++ b/paddlespeech/s2t/training/extensions/evaluator.py @@ -23,7 +23,7 @@ from ..reporter import DictSummary from ..reporter import ObsScope from ..reporter import report from ..timer import Timer -from deepspeech.utils.log import Log +from paddlespeech.s2t.utils.log import Log logger = Log(__name__).getlog() diff --git a/deepspeech/training/extensions/extension.py b/paddlespeech/s2t/training/extensions/extension.py similarity index 100% rename from deepspeech/training/extensions/extension.py rename to paddlespeech/s2t/training/extensions/extension.py diff --git a/deepspeech/training/extensions/plot.py b/paddlespeech/s2t/training/extensions/plot.py similarity index 100% rename from deepspeech/training/extensions/plot.py rename to paddlespeech/s2t/training/extensions/plot.py diff --git a/deepspeech/training/extensions/snapshot.py b/paddlespeech/s2t/training/extensions/snapshot.py similarity index 97% rename from deepspeech/training/extensions/snapshot.py rename to paddlespeech/s2t/training/extensions/snapshot.py index e81eb97f..753c7ec0 100644 --- a/deepspeech/training/extensions/snapshot.py +++ b/paddlespeech/s2t/training/extensions/snapshot.py @@ -23,8 +23,8 @@ import jsonlines from . import extension from ..reporter import get_observations from ..updaters.trainer import Trainer -from deepspeech.utils.log import Log -from deepspeech.utils.mp_tools import rank_zero_only +from paddlespeech.s2t.utils.log import Log +from paddlespeech.s2t.utils.mp_tools import rank_zero_only logger = Log(__name__).getlog() diff --git a/deepspeech/training/extensions/visualizer.py b/paddlespeech/s2t/training/extensions/visualizer.py similarity index 100% rename from deepspeech/training/extensions/visualizer.py rename to paddlespeech/s2t/training/extensions/visualizer.py diff --git a/deepspeech/training/gradclip.py b/paddlespeech/s2t/training/gradclip.py similarity index 98% rename from deepspeech/training/gradclip.py rename to paddlespeech/s2t/training/gradclip.py index 87b36aca..26ac501e 100644 --- a/deepspeech/training/gradclip.py +++ b/paddlespeech/s2t/training/gradclip.py @@ -16,7 +16,7 @@ from paddle.fluid import core from paddle.fluid import layers from paddle.fluid.dygraph import base as imperative_base -from deepspeech.utils.log import Log +from paddlespeech.s2t.utils.log import Log __all__ = ["ClipGradByGlobalNormWithLog"] diff --git a/deepspeech/training/optimizer.py b/paddlespeech/s2t/training/optimizer.py similarity index 93% rename from deepspeech/training/optimizer.py rename to paddlespeech/s2t/training/optimizer.py index db7069c9..8ed30769 100644 --- a/deepspeech/training/optimizer.py +++ b/paddlespeech/s2t/training/optimizer.py @@ -19,10 +19,10 @@ import paddle from paddle.optimizer import Optimizer from paddle.regularizer import L2Decay -from deepspeech.training.gradclip import ClipGradByGlobalNormWithLog -from deepspeech.utils.dynamic_import import dynamic_import -from deepspeech.utils.dynamic_import import instance_class -from deepspeech.utils.log import Log +from paddlespeech.s2t.training.gradclip import ClipGradByGlobalNormWithLog +from paddlespeech.s2t.utils.dynamic_import import dynamic_import +from paddlespeech.s2t.utils.dynamic_import import instance_class +from paddlespeech.s2t.utils.log import Log __all__ = ["OptimizerFactory"] diff --git a/deepspeech/training/reporter.py b/paddlespeech/s2t/training/reporter.py similarity index 100% rename from deepspeech/training/reporter.py rename to paddlespeech/s2t/training/reporter.py diff --git a/deepspeech/training/scheduler.py b/paddlespeech/s2t/training/scheduler.py similarity index 95% rename from deepspeech/training/scheduler.py rename to paddlespeech/s2t/training/scheduler.py index bb53281a..f850f6f5 100644 --- a/deepspeech/training/scheduler.py +++ b/paddlespeech/s2t/training/scheduler.py @@ -19,9 +19,9 @@ from typing import Union from paddle.optimizer.lr import LRScheduler from typeguard import check_argument_types -from deepspeech.utils.dynamic_import import dynamic_import -from deepspeech.utils.dynamic_import import instance_class -from deepspeech.utils.log import Log +from paddlespeech.s2t.utils.dynamic_import import dynamic_import +from paddlespeech.s2t.utils.dynamic_import import instance_class +from paddlespeech.s2t.utils.log import Log __all__ = ["WarmupLR", "LRSchedulerFactory"] diff --git a/deepspeech/training/timer.py b/paddlespeech/s2t/training/timer.py similarity index 97% rename from deepspeech/training/timer.py rename to paddlespeech/s2t/training/timer.py index 2ca9d638..271ffff1 100644 --- a/deepspeech/training/timer.py +++ b/paddlespeech/s2t/training/timer.py @@ -14,7 +14,7 @@ import datetime import time -from deepspeech.utils.log import Log +from paddlespeech.s2t.utils.log import Log __all__ = ["Timer"] diff --git a/deepspeech/training/trainer.py b/paddlespeech/s2t/training/trainer.py similarity index 97% rename from deepspeech/training/trainer.py rename to paddlespeech/s2t/training/trainer.py index 71a27ed2..e6328cdf 100644 --- a/deepspeech/training/trainer.py +++ b/paddlespeech/s2t/training/trainer.py @@ -21,16 +21,16 @@ import paddle from paddle import distributed as dist from tensorboardX import SummaryWriter -from deepspeech.training.reporter import ObsScope -from deepspeech.training.reporter import report -from deepspeech.training.timer import Timer -from deepspeech.utils import mp_tools -from deepspeech.utils import profiler -from deepspeech.utils.checkpoint import Checkpoint -from deepspeech.utils.log import Log -from deepspeech.utils.utility import all_version -from deepspeech.utils.utility import seed_all -from deepspeech.utils.utility import UpdateConfig +from paddlespeech.s2t.training.reporter import ObsScope +from paddlespeech.s2t.training.reporter import report +from paddlespeech.s2t.training.timer import Timer +from paddlespeech.s2t.utils import mp_tools +from paddlespeech.s2t.utils import profiler +from paddlespeech.s2t.utils.checkpoint import Checkpoint +from paddlespeech.s2t.utils.log import Log +from paddlespeech.s2t.utils.utility import all_version +from paddlespeech.s2t.utils.utility import seed_all +from paddlespeech.s2t.utils.utility import UpdateConfig __all__ = ["Trainer"] diff --git a/deepspeech/training/triggers/__init__.py b/paddlespeech/s2t/training/triggers/__init__.py similarity index 100% rename from deepspeech/training/triggers/__init__.py rename to paddlespeech/s2t/training/triggers/__init__.py diff --git a/deepspeech/training/triggers/compare_value_trigger.py b/paddlespeech/s2t/training/triggers/compare_value_trigger.py similarity index 100% rename from deepspeech/training/triggers/compare_value_trigger.py rename to paddlespeech/s2t/training/triggers/compare_value_trigger.py diff --git a/deepspeech/training/triggers/interval_trigger.py b/paddlespeech/s2t/training/triggers/interval_trigger.py similarity index 100% rename from deepspeech/training/triggers/interval_trigger.py rename to paddlespeech/s2t/training/triggers/interval_trigger.py diff --git a/deepspeech/training/triggers/limit_trigger.py b/paddlespeech/s2t/training/triggers/limit_trigger.py similarity index 100% rename from deepspeech/training/triggers/limit_trigger.py rename to paddlespeech/s2t/training/triggers/limit_trigger.py diff --git a/deepspeech/training/triggers/time_trigger.py b/paddlespeech/s2t/training/triggers/time_trigger.py similarity index 100% rename from deepspeech/training/triggers/time_trigger.py rename to paddlespeech/s2t/training/triggers/time_trigger.py diff --git a/deepspeech/training/triggers/utils.py b/paddlespeech/s2t/training/triggers/utils.py similarity index 100% rename from deepspeech/training/triggers/utils.py rename to paddlespeech/s2t/training/triggers/utils.py diff --git a/deepspeech/training/updaters/__init__.py b/paddlespeech/s2t/training/updaters/__init__.py similarity index 100% rename from deepspeech/training/updaters/__init__.py rename to paddlespeech/s2t/training/updaters/__init__.py diff --git a/deepspeech/training/updaters/standard_updater.py b/paddlespeech/s2t/training/updaters/standard_updater.py similarity index 97% rename from deepspeech/training/updaters/standard_updater.py rename to paddlespeech/s2t/training/updaters/standard_updater.py index 10c99e7f..83180a38 100644 --- a/deepspeech/training/updaters/standard_updater.py +++ b/paddlespeech/s2t/training/updaters/standard_updater.py @@ -21,10 +21,10 @@ from paddle.nn import Layer from paddle.optimizer import Optimizer from paddle.optimizer.lr import LRScheduler -from deepspeech.training.reporter import report -from deepspeech.training.updaters.updater import UpdaterBase -from deepspeech.training.updaters.updater import UpdaterState -from deepspeech.utils.log import Log +from paddlespeech.s2t.training.reporter import report +from paddlespeech.s2t.training.updaters.updater import UpdaterBase +from paddlespeech.s2t.training.updaters.updater import UpdaterState +from paddlespeech.s2t.utils.log import Log __all__ = ["StandardUpdater"] diff --git a/deepspeech/training/updaters/trainer.py b/paddlespeech/s2t/training/updaters/trainer.py similarity index 94% rename from deepspeech/training/updaters/trainer.py rename to paddlespeech/s2t/training/updaters/trainer.py index 07769465..dcf844e3 100644 --- a/deepspeech/training/updaters/trainer.py +++ b/paddlespeech/s2t/training/updaters/trainer.py @@ -22,12 +22,12 @@ from typing import Union import six import tqdm -from deepspeech.training.extensions.extension import Extension -from deepspeech.training.extensions.extension import PRIORITY_READER -from deepspeech.training.reporter import ObsScope -from deepspeech.training.triggers import get_trigger -from deepspeech.training.triggers.limit_trigger import LimitTrigger -from deepspeech.training.updaters.updater import UpdaterBase +from paddlespeech.s2t.training.extensions.extension import Extension +from paddlespeech.s2t.training.extensions.extension import PRIORITY_READER +from paddlespeech.s2t.training.reporter import ObsScope +from paddlespeech.s2t.training.triggers import get_trigger +from paddlespeech.s2t.training.triggers.limit_trigger import LimitTrigger +from paddlespeech.s2t.training.updaters.updater import UpdaterBase class _ExtensionEntry(): diff --git a/deepspeech/training/updaters/updater.py b/paddlespeech/s2t/training/updaters/updater.py similarity index 98% rename from deepspeech/training/updaters/updater.py rename to paddlespeech/s2t/training/updaters/updater.py index e5dd6556..ff8d925f 100644 --- a/deepspeech/training/updaters/updater.py +++ b/paddlespeech/s2t/training/updaters/updater.py @@ -15,7 +15,7 @@ from dataclasses import dataclass import paddle -from deepspeech.utils.log import Log +from paddlespeech.s2t.utils.log import Log __all__ = ["UpdaterBase", "UpdaterState"] diff --git a/deepspeech/transform/__init__.py b/paddlespeech/s2t/transform/__init__.py similarity index 100% rename from deepspeech/transform/__init__.py rename to paddlespeech/s2t/transform/__init__.py diff --git a/deepspeech/transform/add_deltas.py b/paddlespeech/s2t/transform/add_deltas.py similarity index 100% rename from deepspeech/transform/add_deltas.py rename to paddlespeech/s2t/transform/add_deltas.py diff --git a/deepspeech/transform/channel_selector.py b/paddlespeech/s2t/transform/channel_selector.py similarity index 100% rename from deepspeech/transform/channel_selector.py rename to paddlespeech/s2t/transform/channel_selector.py diff --git a/deepspeech/transform/cmvn.py b/paddlespeech/s2t/transform/cmvn.py similarity index 100% rename from deepspeech/transform/cmvn.py rename to paddlespeech/s2t/transform/cmvn.py diff --git a/deepspeech/transform/functional.py b/paddlespeech/s2t/transform/functional.py similarity index 94% rename from deepspeech/transform/functional.py rename to paddlespeech/s2t/transform/functional.py index 914e484e..e23d55f9 100644 --- a/deepspeech/transform/functional.py +++ b/paddlespeech/s2t/transform/functional.py @@ -13,8 +13,8 @@ # limitations under the License. import inspect -from deepspeech.transform.transform_interface import TransformInterface -from deepspeech.utils.check_kwargs import check_kwargs +from paddlespeech.s2t.transform.transform_interface import TransformInterface +from paddlespeech.s2t.utils.check_kwargs import check_kwargs class FuncTrans(TransformInterface): diff --git a/deepspeech/transform/perturb.py b/paddlespeech/s2t/transform/perturb.py similarity index 99% rename from deepspeech/transform/perturb.py rename to paddlespeech/s2t/transform/perturb.py index e425fd2e..f7e8bb76 100644 --- a/deepspeech/transform/perturb.py +++ b/paddlespeech/s2t/transform/perturb.py @@ -16,7 +16,7 @@ import numpy import scipy import soundfile -from deepspeech.io.reader import SoundHDF5File +from paddlespeech.s2t.io.reader import SoundHDF5File class SpeedPerturbation(): diff --git a/deepspeech/transform/spec_augment.py b/paddlespeech/s2t/transform/spec_augment.py similarity index 99% rename from deepspeech/transform/spec_augment.py rename to paddlespeech/s2t/transform/spec_augment.py index 0e5324e7..01ead8ad 100644 --- a/deepspeech/transform/spec_augment.py +++ b/paddlespeech/s2t/transform/spec_augment.py @@ -18,7 +18,7 @@ import numpy from PIL import Image from PIL.Image import BICUBIC -from deepspeech.transform.functional import FuncTrans +from paddlespeech.s2t.transform.functional import FuncTrans def time_warp(x, max_time_warp=80, inplace=False, mode="PIL"): diff --git a/deepspeech/transform/spectrogram.py b/paddlespeech/s2t/transform/spectrogram.py similarity index 100% rename from deepspeech/transform/spectrogram.py rename to paddlespeech/s2t/transform/spectrogram.py diff --git a/deepspeech/transform/transform_interface.py b/paddlespeech/s2t/transform/transform_interface.py similarity index 100% rename from deepspeech/transform/transform_interface.py rename to paddlespeech/s2t/transform/transform_interface.py diff --git a/deepspeech/transform/transformation.py b/paddlespeech/s2t/transform/transformation.py similarity index 78% rename from deepspeech/transform/transformation.py rename to paddlespeech/s2t/transform/transformation.py index afb1db28..906048d7 100644 --- a/deepspeech/transform/transformation.py +++ b/paddlespeech/s2t/transform/transformation.py @@ -21,31 +21,31 @@ from inspect import signature import yaml -from deepspeech.utils.dynamic_import import dynamic_import +from paddlespeech.s2t.utils.dynamic_import import dynamic_import # TODO(karita): inherit TransformInterface # TODO(karita): register cmd arguments in asr_train.py import_alias = dict( - identity="deepspeech.transform.transform_interface:Identity", - time_warp="deepspeech.transform.spec_augment:TimeWarp", - time_mask="deepspeech.transform.spec_augment:TimeMask", - freq_mask="deepspeech.transform.spec_augment:FreqMask", - spec_augment="deepspeech.transform.spec_augment:SpecAugment", - speed_perturbation="deepspeech.transform.perturb:SpeedPerturbation", - volume_perturbation="deepspeech.transform.perturb:VolumePerturbation", - noise_injection="deepspeech.transform.perturb:NoiseInjection", - bandpass_perturbation="deepspeech.transform.perturb:BandpassPerturbation", - rir_convolve="deepspeech.transform.perturb:RIRConvolve", - delta="deepspeech.transform.add_deltas:AddDeltas", - cmvn="deepspeech.transform.cmvn:CMVN", - utterance_cmvn="deepspeech.transform.cmvn:UtteranceCMVN", - fbank="deepspeech.transform.spectrogram:LogMelSpectrogram", - spectrogram="deepspeech.transform.spectrogram:Spectrogram", - stft="deepspeech.transform.spectrogram:Stft", - istft="deepspeech.transform.spectrogram:IStft", - stft2fbank="deepspeech.transform.spectrogram:Stft2LogMelSpectrogram", - wpe="deepspeech.transform.wpe:WPE", - channel_selector="deepspeech.transform.channel_selector:ChannelSelector", ) + identity="paddlespeech.s2t.transform.transform_interface:Identity", + time_warp="paddlespeech.s2t.transform.spec_augment:TimeWarp", + time_mask="paddlespeech.s2t.transform.spec_augment:TimeMask", + freq_mask="paddlespeech.s2t.transform.spec_augment:FreqMask", + spec_augment="paddlespeech.s2t.transform.spec_augment:SpecAugment", + speed_perturbation="paddlespeech.s2t.transform.perturb:SpeedPerturbation", + volume_perturbation="paddlespeech.s2t.transform.perturb:VolumePerturbation", + noise_injection="paddlespeech.s2t.transform.perturb:NoiseInjection", + bandpass_perturbation="paddlespeech.s2t.transform.perturb:BandpassPerturbation", + rir_convolve="paddlespeech.s2t.transform.perturb:RIRConvolve", + delta="paddlespeech.s2t.transform.add_deltas:AddDeltas", + cmvn="paddlespeech.s2t.transform.cmvn:CMVN", + utterance_cmvn="paddlespeech.s2t.transform.cmvn:UtteranceCMVN", + fbank="paddlespeech.s2t.transform.spectrogram:LogMelSpectrogram", + spectrogram="paddlespeech.s2t.transform.spectrogram:Spectrogram", + stft="paddlespeech.s2t.transform.spectrogram:Stft", + istft="paddlespeech.s2t.transform.spectrogram:IStft", + stft2fbank="paddlespeech.s2t.transform.spectrogram:Stft2LogMelSpectrogram", + wpe="paddlespeech.s2t.transform.wpe:WPE", + channel_selector="paddlespeech.s2t.transform.channel_selector:ChannelSelector", ) class Transformation(): diff --git a/deepspeech/transform/wpe.py b/paddlespeech/s2t/transform/wpe.py similarity index 100% rename from deepspeech/transform/wpe.py rename to paddlespeech/s2t/transform/wpe.py diff --git a/deepspeech/utils/__init__.py b/paddlespeech/s2t/utils/__init__.py similarity index 100% rename from deepspeech/utils/__init__.py rename to paddlespeech/s2t/utils/__init__.py diff --git a/deepspeech/utils/asr_utils.py b/paddlespeech/s2t/utils/asr_utils.py similarity index 100% rename from deepspeech/utils/asr_utils.py rename to paddlespeech/s2t/utils/asr_utils.py diff --git a/deepspeech/utils/bleu_score.py b/paddlespeech/s2t/utils/bleu_score.py similarity index 100% rename from deepspeech/utils/bleu_score.py rename to paddlespeech/s2t/utils/bleu_score.py diff --git a/deepspeech/utils/check_kwargs.py b/paddlespeech/s2t/utils/check_kwargs.py similarity index 100% rename from deepspeech/utils/check_kwargs.py rename to paddlespeech/s2t/utils/check_kwargs.py diff --git a/deepspeech/utils/checkpoint.py b/paddlespeech/s2t/utils/checkpoint.py similarity index 99% rename from deepspeech/utils/checkpoint.py rename to paddlespeech/s2t/utils/checkpoint.py index 796cafe0..5105f95e 100644 --- a/deepspeech/utils/checkpoint.py +++ b/paddlespeech/s2t/utils/checkpoint.py @@ -23,8 +23,8 @@ import paddle from paddle import distributed as dist from paddle.optimizer import Optimizer -from deepspeech.utils import mp_tools -from deepspeech.utils.log import Log +from paddlespeech.s2t.utils import mp_tools +from paddlespeech.s2t.utils.log import Log logger = Log(__name__).getlog() diff --git a/deepspeech/utils/cli_readers.py b/paddlespeech/s2t/utils/cli_readers.py similarity index 99% rename from deepspeech/utils/cli_readers.py rename to paddlespeech/s2t/utils/cli_readers.py index 72aa2bdb..010adb7a 100644 --- a/deepspeech/utils/cli_readers.py +++ b/paddlespeech/s2t/utils/cli_readers.py @@ -19,7 +19,7 @@ import h5py import kaldiio import soundfile -from deepspeech.io.reader import SoundHDF5File +from paddlespeech.s2t.io.reader import SoundHDF5File def file_reader_helper( diff --git a/deepspeech/utils/cli_utils.py b/paddlespeech/s2t/utils/cli_utils.py similarity index 100% rename from deepspeech/utils/cli_utils.py rename to paddlespeech/s2t/utils/cli_utils.py diff --git a/deepspeech/utils/cli_writers.py b/paddlespeech/s2t/utils/cli_writers.py similarity index 98% rename from deepspeech/utils/cli_writers.py rename to paddlespeech/s2t/utils/cli_writers.py index e0737193..5866a82e 100644 --- a/deepspeech/utils/cli_writers.py +++ b/paddlespeech/s2t/utils/cli_writers.py @@ -19,8 +19,8 @@ import kaldiio import numpy import soundfile -from deepspeech.io.reader import SoundHDF5File -from deepspeech.utils.cli_utils import assert_scipy_wav_style +from paddlespeech.s2t.io.reader import SoundHDF5File +from paddlespeech.s2t.utils.cli_utils import assert_scipy_wav_style def file_writer_helper( diff --git a/deepspeech/utils/ctc_utils.py b/paddlespeech/s2t/utils/ctc_utils.py similarity index 98% rename from deepspeech/utils/ctc_utils.py rename to paddlespeech/s2t/utils/ctc_utils.py index 7e8629c2..ff23350a 100644 --- a/deepspeech/utils/ctc_utils.py +++ b/paddlespeech/s2t/utils/ctc_utils.py @@ -16,9 +16,9 @@ from typing import List import numpy as np import paddle -from deepspeech.utils import text_grid -from deepspeech.utils import utility -from deepspeech.utils.log import Log +from paddlespeech.s2t.utils import text_grid +from paddlespeech.s2t.utils import utility +from paddlespeech.s2t.utils.log import Log logger = Log(__name__).getlog() diff --git a/deepspeech/utils/dynamic_import.py b/paddlespeech/s2t/utils/dynamic_import.py similarity index 89% rename from deepspeech/utils/dynamic_import.py rename to paddlespeech/s2t/utils/dynamic_import.py index 533f15ee..3eb0db14 100644 --- a/deepspeech/utils/dynamic_import.py +++ b/paddlespeech/s2t/utils/dynamic_import.py @@ -18,8 +18,8 @@ from typing import Dict from typing import List from typing import Text -from deepspeech.utils.log import Log -from deepspeech.utils.tensor_utils import has_tensor +from paddlespeech.s2t.utils.log import Log +from paddlespeech.s2t.utils.tensor_utils import has_tensor logger = Log(__name__).getlog() @@ -30,13 +30,13 @@ def dynamic_import(import_path, alias=dict()): """dynamic import module and class :param str import_path: syntax 'module_name:class_name' - e.g., 'deepspeech.models.u2:U2Model' + e.g., 'paddlespeech.s2t.models.u2:U2Model' :param dict alias: shortcut for registered class :return: imported class """ if import_path not in alias and ":" not in import_path: raise ValueError("import_path should be one of {} or " - 'include ":", e.g. "deepspeech.models.u2:U2Model" : ' + 'include ":", e.g. "paddlespeech.s2t.models.u2:U2Model" : ' "{}".format(set(alias), import_path)) if ":" not in import_path: import_path = alias[import_path] diff --git a/deepspeech/utils/error_rate.py b/paddlespeech/s2t/utils/error_rate.py similarity index 100% rename from deepspeech/utils/error_rate.py rename to paddlespeech/s2t/utils/error_rate.py diff --git a/deepspeech/utils/layer_tools.py b/paddlespeech/s2t/utils/layer_tools.py similarity index 100% rename from deepspeech/utils/layer_tools.py rename to paddlespeech/s2t/utils/layer_tools.py diff --git a/deepspeech/utils/log.py b/paddlespeech/s2t/utils/log.py similarity index 100% rename from deepspeech/utils/log.py rename to paddlespeech/s2t/utils/log.py diff --git a/deepspeech/utils/mp_tools.py b/paddlespeech/s2t/utils/mp_tools.py similarity index 100% rename from deepspeech/utils/mp_tools.py rename to paddlespeech/s2t/utils/mp_tools.py diff --git a/deepspeech/utils/profiler.py b/paddlespeech/s2t/utils/profiler.py similarity index 99% rename from deepspeech/utils/profiler.py rename to paddlespeech/s2t/utils/profiler.py index 5733f8ed..3592157d 100644 --- a/deepspeech/utils/profiler.py +++ b/paddlespeech/s2t/utils/profiler.py @@ -15,7 +15,7 @@ import sys import paddle -from deepspeech.utils.log import Log +from paddlespeech.s2t.utils.log import Log logger = Log(__name__).getlog() diff --git a/deepspeech/utils/socket_server.py b/paddlespeech/s2t/utils/socket_server.py similarity index 98% rename from deepspeech/utils/socket_server.py rename to paddlespeech/s2t/utils/socket_server.py index 45c659f6..43b56d72 100644 --- a/deepspeech/utils/socket_server.py +++ b/paddlespeech/s2t/utils/socket_server.py @@ -21,7 +21,7 @@ import wave from time import gmtime from time import strftime -from deepspeech.frontend.utility import read_manifest +from paddlespeech.s2t.frontend.utility import read_manifest __all__ = ["socket_send", "warm_up_test", "AsrTCPServer", "AsrRequestHandler"] diff --git a/deepspeech/utils/spec_augment.py b/paddlespeech/s2t/utils/spec_augment.py similarity index 100% rename from deepspeech/utils/spec_augment.py rename to paddlespeech/s2t/utils/spec_augment.py diff --git a/deepspeech/utils/tensor_utils.py b/paddlespeech/s2t/utils/tensor_utils.py similarity index 99% rename from deepspeech/utils/tensor_utils.py rename to paddlespeech/s2t/utils/tensor_utils.py index 0cc03b19..0dbaa0b6 100644 --- a/deepspeech/utils/tensor_utils.py +++ b/paddlespeech/s2t/utils/tensor_utils.py @@ -17,7 +17,7 @@ from typing import Tuple import paddle -from deepspeech.utils.log import Log +from paddlespeech.s2t.utils.log import Log __all__ = ["pad_sequence", "add_sos_eos", "th_accuracy", "has_tensor"] diff --git a/deepspeech/utils/text_grid.py b/paddlespeech/s2t/utils/text_grid.py similarity index 100% rename from deepspeech/utils/text_grid.py rename to paddlespeech/s2t/utils/text_grid.py diff --git a/deepspeech/utils/utility.py b/paddlespeech/s2t/utils/utility.py similarity index 98% rename from deepspeech/utils/utility.py rename to paddlespeech/s2t/utils/utility.py index 8773b84c..73c79816 100644 --- a/deepspeech/utils/utility.py +++ b/paddlespeech/s2t/utils/utility.py @@ -25,7 +25,7 @@ import numpy as np import paddle import soundfile -from deepspeech.utils.log import Log +from paddlespeech.s2t.utils.log import Log logger = Log(__name__).getlog() diff --git a/parakeet/__init__.py b/paddlespeech/t2s/__init__.py similarity index 100% rename from parakeet/__init__.py rename to paddlespeech/t2s/__init__.py diff --git a/parakeet/audio/__init__.py b/paddlespeech/t2s/audio/__init__.py similarity index 100% rename from parakeet/audio/__init__.py rename to paddlespeech/t2s/audio/__init__.py diff --git a/parakeet/audio/audio.py b/paddlespeech/t2s/audio/audio.py similarity index 100% rename from parakeet/audio/audio.py rename to paddlespeech/t2s/audio/audio.py diff --git a/parakeet/audio/spec_normalizer.py b/paddlespeech/t2s/audio/spec_normalizer.py similarity index 100% rename from parakeet/audio/spec_normalizer.py rename to paddlespeech/t2s/audio/spec_normalizer.py diff --git a/parakeet/data/__init__.py b/paddlespeech/t2s/data/__init__.py similarity index 92% rename from parakeet/data/__init__.py rename to paddlespeech/t2s/data/__init__.py index c71c05bd..c605205d 100644 --- a/parakeet/data/__init__.py +++ b/paddlespeech/t2s/data/__init__.py @@ -11,7 +11,7 @@ # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. -"""Parakeet's infrastructure for data processing. +"""t2s's infrastructure for data processing. """ from .batch import * from .dataset import * diff --git a/parakeet/data/batch.py b/paddlespeech/t2s/data/batch.py similarity index 100% rename from parakeet/data/batch.py rename to paddlespeech/t2s/data/batch.py diff --git a/parakeet/data/dataset.py b/paddlespeech/t2s/data/dataset.py similarity index 100% rename from parakeet/data/dataset.py rename to paddlespeech/t2s/data/dataset.py diff --git a/parakeet/data/get_feats.py b/paddlespeech/t2s/data/get_feats.py similarity index 100% rename from parakeet/data/get_feats.py rename to paddlespeech/t2s/data/get_feats.py diff --git a/parakeet/datasets/__init__.py b/paddlespeech/t2s/datasets/__init__.py similarity index 100% rename from parakeet/datasets/__init__.py rename to paddlespeech/t2s/datasets/__init__.py diff --git a/parakeet/datasets/am_batch_fn.py b/paddlespeech/t2s/datasets/am_batch_fn.py similarity index 99% rename from parakeet/datasets/am_batch_fn.py rename to paddlespeech/t2s/datasets/am_batch_fn.py index 42d5babd..5ed9aa7a 100644 --- a/parakeet/datasets/am_batch_fn.py +++ b/paddlespeech/t2s/datasets/am_batch_fn.py @@ -14,7 +14,7 @@ import numpy as np import paddle -from parakeet.data.batch import batch_sequences +from paddlespeech.t2s.data.batch import batch_sequences def speedyspeech_batch_fn(examples): diff --git a/parakeet/datasets/common.py b/paddlespeech/t2s/datasets/common.py similarity index 100% rename from parakeet/datasets/common.py rename to paddlespeech/t2s/datasets/common.py diff --git a/parakeet/datasets/data_table.py b/paddlespeech/t2s/datasets/data_table.py similarity index 100% rename from parakeet/datasets/data_table.py rename to paddlespeech/t2s/datasets/data_table.py diff --git a/parakeet/datasets/ljspeech.py b/paddlespeech/t2s/datasets/ljspeech.py similarity index 100% rename from parakeet/datasets/ljspeech.py rename to paddlespeech/t2s/datasets/ljspeech.py diff --git a/parakeet/datasets/preprocess_utils.py b/paddlespeech/t2s/datasets/preprocess_utils.py similarity index 100% rename from parakeet/datasets/preprocess_utils.py rename to paddlespeech/t2s/datasets/preprocess_utils.py diff --git a/parakeet/datasets/vocoder_batch_fn.py b/paddlespeech/t2s/datasets/vocoder_batch_fn.py similarity index 100% rename from parakeet/datasets/vocoder_batch_fn.py rename to paddlespeech/t2s/datasets/vocoder_batch_fn.py diff --git a/parakeet/exps/__init__.py b/paddlespeech/t2s/exps/__init__.py similarity index 100% rename from parakeet/exps/__init__.py rename to paddlespeech/t2s/exps/__init__.py diff --git a/parakeet/exps/fastspeech2/__init__.py b/paddlespeech/t2s/exps/fastspeech2/__init__.py similarity index 100% rename from parakeet/exps/fastspeech2/__init__.py rename to paddlespeech/t2s/exps/fastspeech2/__init__.py diff --git a/parakeet/exps/fastspeech2/inference.py b/paddlespeech/t2s/exps/fastspeech2/inference.py similarity index 98% rename from parakeet/exps/fastspeech2/inference.py rename to paddlespeech/t2s/exps/fastspeech2/inference.py index 43676088..07e9ed7e 100644 --- a/parakeet/exps/fastspeech2/inference.py +++ b/paddlespeech/t2s/exps/fastspeech2/inference.py @@ -18,7 +18,7 @@ from pathlib import Path import soundfile as sf from paddle import inference -from parakeet.frontend.zh_frontend import Frontend +from paddlespeech.t2s.frontend.zh_frontend import Frontend def main(): diff --git a/parakeet/exps/fastspeech2/multi_spk_synthesize_e2e.py b/paddlespeech/t2s/exps/fastspeech2/multi_spk_synthesize_e2e.py similarity index 93% rename from parakeet/exps/fastspeech2/multi_spk_synthesize_e2e.py rename to paddlespeech/t2s/exps/fastspeech2/multi_spk_synthesize_e2e.py index 825b3ed3..98cf9f8f 100644 --- a/parakeet/exps/fastspeech2/multi_spk_synthesize_e2e.py +++ b/paddlespeech/t2s/exps/fastspeech2/multi_spk_synthesize_e2e.py @@ -21,12 +21,12 @@ import soundfile as sf import yaml from yacs.config import CfgNode -from parakeet.frontend.zh_frontend import Frontend -from parakeet.models.fastspeech2 import FastSpeech2 -from parakeet.models.fastspeech2 import FastSpeech2Inference -from parakeet.models.parallel_wavegan import PWGGenerator -from parakeet.models.parallel_wavegan import PWGInference -from parakeet.modules.normalizer import ZScore +from paddlespeech.t2s.frontend.zh_frontend import Frontend +from paddlespeech.t2s.models.fastspeech2 import FastSpeech2 +from paddlespeech.t2s.models.fastspeech2 import FastSpeech2Inference +from paddlespeech.t2s.models.parallel_wavegan import PWGGenerator +from paddlespeech.t2s.models.parallel_wavegan import PWGInference +from paddlespeech.t2s.modules.normalizer import ZScore def evaluate(args, fastspeech2_config, pwg_config): diff --git a/parakeet/exps/fastspeech2/multi_spk_synthesize_e2e_en.py b/paddlespeech/t2s/exps/fastspeech2/multi_spk_synthesize_e2e_en.py similarity index 94% rename from parakeet/exps/fastspeech2/multi_spk_synthesize_e2e_en.py rename to paddlespeech/t2s/exps/fastspeech2/multi_spk_synthesize_e2e_en.py index a4761982..9e29eea1 100644 --- a/parakeet/exps/fastspeech2/multi_spk_synthesize_e2e_en.py +++ b/paddlespeech/t2s/exps/fastspeech2/multi_spk_synthesize_e2e_en.py @@ -21,12 +21,12 @@ import soundfile as sf import yaml from yacs.config import CfgNode -from parakeet.frontend import English -from parakeet.models.fastspeech2 import FastSpeech2 -from parakeet.models.fastspeech2 import FastSpeech2Inference -from parakeet.models.parallel_wavegan import PWGGenerator -from parakeet.models.parallel_wavegan import PWGInference -from parakeet.modules.normalizer import ZScore +from paddlespeech.t2s.frontend import English +from paddlespeech.t2s.models.fastspeech2 import FastSpeech2 +from paddlespeech.t2s.models.fastspeech2 import FastSpeech2Inference +from paddlespeech.t2s.models.parallel_wavegan import PWGGenerator +from paddlespeech.t2s.models.parallel_wavegan import PWGInference +from paddlespeech.t2s.modules.normalizer import ZScore def evaluate(args, fastspeech2_config, pwg_config): diff --git a/parakeet/exps/fastspeech2/normalize.py b/paddlespeech/t2s/exps/fastspeech2/normalize.py similarity index 99% rename from parakeet/exps/fastspeech2/normalize.py rename to paddlespeech/t2s/exps/fastspeech2/normalize.py index b4b31e31..7283f6b4 100644 --- a/parakeet/exps/fastspeech2/normalize.py +++ b/paddlespeech/t2s/exps/fastspeech2/normalize.py @@ -22,7 +22,7 @@ import numpy as np from sklearn.preprocessing import StandardScaler from tqdm import tqdm -from parakeet.datasets.data_table import DataTable +from paddlespeech.t2s.datasets.data_table import DataTable def main(): diff --git a/parakeet/exps/fastspeech2/preprocess.py b/paddlespeech/t2s/exps/fastspeech2/preprocess.py similarity index 95% rename from parakeet/exps/fastspeech2/preprocess.py rename to paddlespeech/t2s/exps/fastspeech2/preprocess.py index bb796b64..3702ecd3 100644 --- a/parakeet/exps/fastspeech2/preprocess.py +++ b/paddlespeech/t2s/exps/fastspeech2/preprocess.py @@ -27,14 +27,14 @@ import tqdm import yaml from yacs.config import CfgNode -from parakeet.data.get_feats import Energy -from parakeet.data.get_feats import LogMelFBank -from parakeet.data.get_feats import Pitch -from parakeet.datasets.preprocess_utils import compare_duration_and_mel_length -from parakeet.datasets.preprocess_utils import get_input_token -from parakeet.datasets.preprocess_utils import get_phn_dur -from parakeet.datasets.preprocess_utils import get_spk_id_map -from parakeet.datasets.preprocess_utils import merge_silence +from paddlespeech.t2s.data.get_feats import Energy +from paddlespeech.t2s.data.get_feats import LogMelFBank +from paddlespeech.t2s.data.get_feats import Pitch +from paddlespeech.t2s.datasets.preprocess_utils import compare_duration_and_mel_length +from paddlespeech.t2s.datasets.preprocess_utils import get_input_token +from paddlespeech.t2s.datasets.preprocess_utils import get_phn_dur +from paddlespeech.t2s.datasets.preprocess_utils import get_spk_id_map +from paddlespeech.t2s.datasets.preprocess_utils import merge_silence def process_sentence(config: Dict[str, Any], diff --git a/parakeet/exps/fastspeech2/synthesize.py b/paddlespeech/t2s/exps/fastspeech2/synthesize.py similarity index 93% rename from parakeet/exps/fastspeech2/synthesize.py rename to paddlespeech/t2s/exps/fastspeech2/synthesize.py index 91327757..1beac5ce 100644 --- a/parakeet/exps/fastspeech2/synthesize.py +++ b/paddlespeech/t2s/exps/fastspeech2/synthesize.py @@ -22,12 +22,12 @@ import soundfile as sf import yaml from yacs.config import CfgNode -from parakeet.datasets.data_table import DataTable -from parakeet.models.fastspeech2 import FastSpeech2 -from parakeet.models.fastspeech2 import FastSpeech2Inference -from parakeet.models.parallel_wavegan import PWGGenerator -from parakeet.models.parallel_wavegan import PWGInference -from parakeet.modules.normalizer import ZScore +from paddlespeech.t2s.datasets.data_table import DataTable +from paddlespeech.t2s.models.fastspeech2 import FastSpeech2 +from paddlespeech.t2s.models.fastspeech2 import FastSpeech2Inference +from paddlespeech.t2s.models.parallel_wavegan import PWGGenerator +from paddlespeech.t2s.models.parallel_wavegan import PWGInference +from paddlespeech.t2s.modules.normalizer import ZScore def evaluate(args, fastspeech2_config, pwg_config): diff --git a/parakeet/exps/fastspeech2/synthesize_e2e.py b/paddlespeech/t2s/exps/fastspeech2/synthesize_e2e.py similarity index 94% rename from parakeet/exps/fastspeech2/synthesize_e2e.py rename to paddlespeech/t2s/exps/fastspeech2/synthesize_e2e.py index 9c036e9f..b6a8fc58 100644 --- a/parakeet/exps/fastspeech2/synthesize_e2e.py +++ b/paddlespeech/t2s/exps/fastspeech2/synthesize_e2e.py @@ -24,12 +24,12 @@ from paddle import jit from paddle.static import InputSpec from yacs.config import CfgNode -from parakeet.frontend.zh_frontend import Frontend -from parakeet.models.fastspeech2 import FastSpeech2 -from parakeet.models.fastspeech2 import FastSpeech2Inference -from parakeet.models.parallel_wavegan import PWGGenerator -from parakeet.models.parallel_wavegan import PWGInference -from parakeet.modules.normalizer import ZScore +from paddlespeech.t2s.frontend.zh_frontend import Frontend +from paddlespeech.t2s.models.fastspeech2 import FastSpeech2 +from paddlespeech.t2s.models.fastspeech2 import FastSpeech2Inference +from paddlespeech.t2s.models.parallel_wavegan import PWGGenerator +from paddlespeech.t2s.models.parallel_wavegan import PWGInference +from paddlespeech.t2s.modules.normalizer import ZScore def evaluate(args, fastspeech2_config, pwg_config): diff --git a/parakeet/exps/fastspeech2/synthesize_e2e_en.py b/paddlespeech/t2s/exps/fastspeech2/synthesize_e2e_en.py similarity index 93% rename from parakeet/exps/fastspeech2/synthesize_e2e_en.py rename to paddlespeech/t2s/exps/fastspeech2/synthesize_e2e_en.py index 4e8a20c7..7a55fbb1 100644 --- a/parakeet/exps/fastspeech2/synthesize_e2e_en.py +++ b/paddlespeech/t2s/exps/fastspeech2/synthesize_e2e_en.py @@ -21,12 +21,12 @@ import soundfile as sf import yaml from yacs.config import CfgNode -from parakeet.frontend import English -from parakeet.models.fastspeech2 import FastSpeech2 -from parakeet.models.fastspeech2 import FastSpeech2Inference -from parakeet.models.parallel_wavegan import PWGGenerator -from parakeet.models.parallel_wavegan import PWGInference -from parakeet.modules.normalizer import ZScore +from paddlespeech.t2s.frontend import English +from paddlespeech.t2s.models.fastspeech2 import FastSpeech2 +from paddlespeech.t2s.models.fastspeech2 import FastSpeech2Inference +from paddlespeech.t2s.models.parallel_wavegan import PWGGenerator +from paddlespeech.t2s.models.parallel_wavegan import PWGInference +from paddlespeech.t2s.modules.normalizer import ZScore def evaluate(args, fastspeech2_config, pwg_config): diff --git a/parakeet/exps/fastspeech2/synthesize_e2e_melgan.py b/paddlespeech/t2s/exps/fastspeech2/synthesize_e2e_melgan.py similarity index 94% rename from parakeet/exps/fastspeech2/synthesize_e2e_melgan.py rename to paddlespeech/t2s/exps/fastspeech2/synthesize_e2e_melgan.py index 9fba9593..92a43d5c 100644 --- a/parakeet/exps/fastspeech2/synthesize_e2e_melgan.py +++ b/paddlespeech/t2s/exps/fastspeech2/synthesize_e2e_melgan.py @@ -24,12 +24,12 @@ from paddle import jit from paddle.static import InputSpec from yacs.config import CfgNode -from parakeet.frontend.zh_frontend import Frontend -from parakeet.models.fastspeech2 import FastSpeech2 -from parakeet.models.fastspeech2 import FastSpeech2Inference -from parakeet.models.melgan import MelGANGenerator -from parakeet.models.melgan import MelGANInference -from parakeet.modules.normalizer import ZScore +from paddlespeech.t2s.frontend.zh_frontend import Frontend +from paddlespeech.t2s.models.fastspeech2 import FastSpeech2 +from paddlespeech.t2s.models.fastspeech2 import FastSpeech2Inference +from paddlespeech.t2s.models.melgan import MelGANGenerator +from paddlespeech.t2s.models.melgan import MelGANInference +from paddlespeech.t2s.modules.normalizer import ZScore def evaluate(args, fastspeech2_config, melgan_config): diff --git a/parakeet/exps/fastspeech2/train.py b/paddlespeech/t2s/exps/fastspeech2/train.py similarity index 89% rename from parakeet/exps/fastspeech2/train.py rename to paddlespeech/t2s/exps/fastspeech2/train.py index 47ad1b4d..5662d15d 100644 --- a/parakeet/exps/fastspeech2/train.py +++ b/paddlespeech/t2s/exps/fastspeech2/train.py @@ -27,17 +27,17 @@ from paddle.io import DataLoader from paddle.io import DistributedBatchSampler from yacs.config import CfgNode -from parakeet.datasets.am_batch_fn import fastspeech2_multi_spk_batch_fn -from parakeet.datasets.am_batch_fn import fastspeech2_single_spk_batch_fn -from parakeet.datasets.data_table import DataTable -from parakeet.models.fastspeech2 import FastSpeech2 -from parakeet.models.fastspeech2 import FastSpeech2Evaluator -from parakeet.models.fastspeech2 import FastSpeech2Updater -from parakeet.training.extensions.snapshot import Snapshot -from parakeet.training.extensions.visualizer import VisualDL -from parakeet.training.optimizer import build_optimizers -from parakeet.training.seeding import seed_everything -from parakeet.training.trainer import Trainer +from paddlespeech.t2s.datasets.am_batch_fn import fastspeech2_multi_spk_batch_fn +from paddlespeech.t2s.datasets.am_batch_fn import fastspeech2_single_spk_batch_fn +from paddlespeech.t2s.datasets.data_table import DataTable +from paddlespeech.t2s.models.fastspeech2 import FastSpeech2 +from paddlespeech.t2s.models.fastspeech2 import FastSpeech2Evaluator +from paddlespeech.t2s.models.fastspeech2 import FastSpeech2Updater +from paddlespeech.t2s.training.extensions.snapshot import Snapshot +from paddlespeech.t2s.training.extensions.visualizer import VisualDL +from paddlespeech.t2s.training.optimizer import build_optimizers +from paddlespeech.t2s.training.seeding import seed_everything +from paddlespeech.t2s.training.trainer import Trainer def train_sp(args, config): diff --git a/parakeet/exps/gan_vocoder/README.md b/paddlespeech/t2s/exps/gan_vocoder/README.md similarity index 100% rename from parakeet/exps/gan_vocoder/README.md rename to paddlespeech/t2s/exps/gan_vocoder/README.md diff --git a/parakeet/exps/gan_vocoder/__init__.py b/paddlespeech/t2s/exps/gan_vocoder/__init__.py similarity index 100% rename from parakeet/exps/gan_vocoder/__init__.py rename to paddlespeech/t2s/exps/gan_vocoder/__init__.py diff --git a/parakeet/exps/gan_vocoder/multi_band_melgan/__init__.py b/paddlespeech/t2s/exps/gan_vocoder/multi_band_melgan/__init__.py similarity index 100% rename from parakeet/exps/gan_vocoder/multi_band_melgan/__init__.py rename to paddlespeech/t2s/exps/gan_vocoder/multi_band_melgan/__init__.py diff --git a/parakeet/exps/gan_vocoder/multi_band_melgan/synthesize.py b/paddlespeech/t2s/exps/gan_vocoder/multi_band_melgan/synthesize.py similarity index 96% rename from parakeet/exps/gan_vocoder/multi_band_melgan/synthesize.py rename to paddlespeech/t2s/exps/gan_vocoder/multi_band_melgan/synthesize.py index 00b1b96c..720b08ce 100644 --- a/parakeet/exps/gan_vocoder/multi_band_melgan/synthesize.py +++ b/paddlespeech/t2s/exps/gan_vocoder/multi_band_melgan/synthesize.py @@ -24,8 +24,8 @@ from paddle import distributed as dist from timer import timer from yacs.config import CfgNode -from parakeet.datasets.data_table import DataTable -from parakeet.models.melgan import MelGANGenerator +from paddlespeech.t2s.datasets.data_table import DataTable +from paddlespeech.t2s.models.melgan import MelGANGenerator def main(): diff --git a/parakeet/exps/gan_vocoder/multi_band_melgan/train.py b/paddlespeech/t2s/exps/gan_vocoder/multi_band_melgan/train.py similarity index 90% rename from parakeet/exps/gan_vocoder/multi_band_melgan/train.py rename to paddlespeech/t2s/exps/gan_vocoder/multi_band_melgan/train.py index c03fb354..45704607 100644 --- a/parakeet/exps/gan_vocoder/multi_band_melgan/train.py +++ b/paddlespeech/t2s/exps/gan_vocoder/multi_band_melgan/train.py @@ -30,20 +30,20 @@ from paddle.optimizer import Adam from paddle.optimizer.lr import MultiStepDecay from yacs.config import CfgNode -from parakeet.datasets.data_table import DataTable -from parakeet.datasets.vocoder_batch_fn import Clip -from parakeet.models.melgan import MBMelGANEvaluator -from parakeet.models.melgan import MBMelGANUpdater -from parakeet.models.melgan import MelGANGenerator -from parakeet.models.melgan import MelGANMultiScaleDiscriminator -from parakeet.modules.adversarial_loss import DiscriminatorAdversarialLoss -from parakeet.modules.adversarial_loss import GeneratorAdversarialLoss -from parakeet.modules.pqmf import PQMF -from parakeet.modules.stft_loss import MultiResolutionSTFTLoss -from parakeet.training.extensions.snapshot import Snapshot -from parakeet.training.extensions.visualizer import VisualDL -from parakeet.training.seeding import seed_everything -from parakeet.training.trainer import Trainer +from paddlespeech.t2s.datasets.data_table import DataTable +from paddlespeech.t2s.datasets.vocoder_batch_fn import Clip +from paddlespeech.t2s.models.melgan import MBMelGANEvaluator +from paddlespeech.t2s.models.melgan import MBMelGANUpdater +from paddlespeech.t2s.models.melgan import MelGANGenerator +from paddlespeech.t2s.models.melgan import MelGANMultiScaleDiscriminator +from paddlespeech.t2s.modules.adversarial_loss import DiscriminatorAdversarialLoss +from paddlespeech.t2s.modules.adversarial_loss import GeneratorAdversarialLoss +from paddlespeech.t2s.modules.pqmf import PQMF +from paddlespeech.t2s.modules.stft_loss import MultiResolutionSTFTLoss +from paddlespeech.t2s.training.extensions.snapshot import Snapshot +from paddlespeech.t2s.training.extensions.visualizer import VisualDL +from paddlespeech.t2s.training.seeding import seed_everything +from paddlespeech.t2s.training.trainer import Trainer def train_sp(args, config): diff --git a/parakeet/exps/gan_vocoder/normalize.py b/paddlespeech/t2s/exps/gan_vocoder/normalize.py similarity index 98% rename from parakeet/exps/gan_vocoder/normalize.py rename to paddlespeech/t2s/exps/gan_vocoder/normalize.py index c772594b..ba95d3ed 100644 --- a/parakeet/exps/gan_vocoder/normalize.py +++ b/paddlespeech/t2s/exps/gan_vocoder/normalize.py @@ -22,7 +22,7 @@ import numpy as np from sklearn.preprocessing import StandardScaler from tqdm import tqdm -from parakeet.datasets.data_table import DataTable +from paddlespeech.t2s.datasets.data_table import DataTable def main(): diff --git a/parakeet/exps/gan_vocoder/parallelwave_gan/__init__.py b/paddlespeech/t2s/exps/gan_vocoder/parallelwave_gan/__init__.py similarity index 100% rename from parakeet/exps/gan_vocoder/parallelwave_gan/__init__.py rename to paddlespeech/t2s/exps/gan_vocoder/parallelwave_gan/__init__.py diff --git a/parakeet/exps/gan_vocoder/parallelwave_gan/synthesize.py b/paddlespeech/t2s/exps/gan_vocoder/parallelwave_gan/synthesize.py similarity index 96% rename from parakeet/exps/gan_vocoder/parallelwave_gan/synthesize.py rename to paddlespeech/t2s/exps/gan_vocoder/parallelwave_gan/synthesize.py index 2400e00b..ce90aaf4 100644 --- a/parakeet/exps/gan_vocoder/parallelwave_gan/synthesize.py +++ b/paddlespeech/t2s/exps/gan_vocoder/parallelwave_gan/synthesize.py @@ -24,8 +24,8 @@ from paddle import distributed as dist from timer import timer from yacs.config import CfgNode -from parakeet.datasets.data_table import DataTable -from parakeet.models.parallel_wavegan import PWGGenerator +from paddlespeech.t2s.datasets.data_table import DataTable +from paddlespeech.t2s.models.parallel_wavegan import PWGGenerator def main(): diff --git a/parakeet/exps/gan_vocoder/parallelwave_gan/synthesize_from_wav.py b/paddlespeech/t2s/exps/gan_vocoder/parallelwave_gan/synthesize_from_wav.py similarity index 93% rename from parakeet/exps/gan_vocoder/parallelwave_gan/synthesize_from_wav.py rename to paddlespeech/t2s/exps/gan_vocoder/parallelwave_gan/synthesize_from_wav.py index c451a51c..a04a547e 100644 --- a/parakeet/exps/gan_vocoder/parallelwave_gan/synthesize_from_wav.py +++ b/paddlespeech/t2s/exps/gan_vocoder/parallelwave_gan/synthesize_from_wav.py @@ -23,10 +23,10 @@ import soundfile as sf import yaml from yacs.config import CfgNode -from parakeet.data.get_feats import LogMelFBank -from parakeet.models.parallel_wavegan import PWGGenerator -from parakeet.models.parallel_wavegan import PWGInference -from parakeet.modules.normalizer import ZScore +from paddlespeech.t2s.data.get_feats import LogMelFBank +from paddlespeech.t2s.models.parallel_wavegan import PWGGenerator +from paddlespeech.t2s.models.parallel_wavegan import PWGInference +from paddlespeech.t2s.modules.normalizer import ZScore def evaluate(args, config): diff --git a/parakeet/exps/gan_vocoder/parallelwave_gan/train.py b/paddlespeech/t2s/exps/gan_vocoder/parallelwave_gan/train.py similarity index 92% rename from parakeet/exps/gan_vocoder/parallelwave_gan/train.py rename to paddlespeech/t2s/exps/gan_vocoder/parallelwave_gan/train.py index ad50b65c..99801267 100644 --- a/parakeet/exps/gan_vocoder/parallelwave_gan/train.py +++ b/paddlespeech/t2s/exps/gan_vocoder/parallelwave_gan/train.py @@ -30,17 +30,17 @@ from paddle.optimizer import Adam # No RAdaom from paddle.optimizer.lr import StepDecay from yacs.config import CfgNode -from parakeet.datasets.data_table import DataTable -from parakeet.datasets.vocoder_batch_fn import Clip -from parakeet.models.parallel_wavegan import PWGDiscriminator -from parakeet.models.parallel_wavegan import PWGEvaluator -from parakeet.models.parallel_wavegan import PWGGenerator -from parakeet.models.parallel_wavegan import PWGUpdater -from parakeet.modules.stft_loss import MultiResolutionSTFTLoss -from parakeet.training.extensions.snapshot import Snapshot -from parakeet.training.extensions.visualizer import VisualDL -from parakeet.training.seeding import seed_everything -from parakeet.training.trainer import Trainer +from paddlespeech.t2s.datasets.data_table import DataTable +from paddlespeech.t2s.datasets.vocoder_batch_fn import Clip +from paddlespeech.t2s.models.parallel_wavegan import PWGDiscriminator +from paddlespeech.t2s.models.parallel_wavegan import PWGEvaluator +from paddlespeech.t2s.models.parallel_wavegan import PWGGenerator +from paddlespeech.t2s.models.parallel_wavegan import PWGUpdater +from paddlespeech.t2s.modules.stft_loss import MultiResolutionSTFTLoss +from paddlespeech.t2s.training.extensions.snapshot import Snapshot +from paddlespeech.t2s.training.extensions.visualizer import VisualDL +from paddlespeech.t2s.training.seeding import seed_everything +from paddlespeech.t2s.training.trainer import Trainer def train_sp(args, config): diff --git a/parakeet/exps/gan_vocoder/preprocess.py b/paddlespeech/t2s/exps/gan_vocoder/preprocess.py similarity index 98% rename from parakeet/exps/gan_vocoder/preprocess.py rename to paddlespeech/t2s/exps/gan_vocoder/preprocess.py index c10143c7..852b0c91 100644 --- a/parakeet/exps/gan_vocoder/preprocess.py +++ b/paddlespeech/t2s/exps/gan_vocoder/preprocess.py @@ -27,9 +27,9 @@ import tqdm import yaml from yacs.config import CfgNode -from parakeet.data.get_feats import LogMelFBank -from parakeet.datasets.preprocess_utils import get_phn_dur -from parakeet.datasets.preprocess_utils import merge_silence +from paddlespeech.t2s.data.get_feats import LogMelFBank +from paddlespeech.t2s.datasets.preprocess_utils import get_phn_dur +from paddlespeech.t2s.datasets.preprocess_utils import merge_silence def process_sentence(config: Dict[str, Any], diff --git a/parakeet/exps/ge2e/__init__.py b/paddlespeech/t2s/exps/ge2e/__init__.py similarity index 100% rename from parakeet/exps/ge2e/__init__.py rename to paddlespeech/t2s/exps/ge2e/__init__.py diff --git a/parakeet/exps/ge2e/audio_processor.py b/paddlespeech/t2s/exps/ge2e/audio_processor.py similarity index 100% rename from parakeet/exps/ge2e/audio_processor.py rename to paddlespeech/t2s/exps/ge2e/audio_processor.py diff --git a/parakeet/exps/ge2e/config.py b/paddlespeech/t2s/exps/ge2e/config.py similarity index 100% rename from parakeet/exps/ge2e/config.py rename to paddlespeech/t2s/exps/ge2e/config.py diff --git a/parakeet/exps/ge2e/dataset_processors.py b/paddlespeech/t2s/exps/ge2e/dataset_processors.py similarity index 98% rename from parakeet/exps/ge2e/dataset_processors.py rename to paddlespeech/t2s/exps/ge2e/dataset_processors.py index 29b58410..a9320d98 100644 --- a/parakeet/exps/ge2e/dataset_processors.py +++ b/paddlespeech/t2s/exps/ge2e/dataset_processors.py @@ -19,7 +19,7 @@ from typing import List import numpy as np from tqdm import tqdm -from parakeet.exps.ge2e.audio_processor import SpeakerVerificationPreprocessor +from paddlespeech.t2s.exps.ge2e.audio_processor import SpeakerVerificationPreprocessor def _process_utterance(path_pair, processor: SpeakerVerificationPreprocessor): diff --git a/parakeet/exps/ge2e/inference.py b/paddlespeech/t2s/exps/ge2e/inference.py similarity index 95% rename from parakeet/exps/ge2e/inference.py rename to paddlespeech/t2s/exps/ge2e/inference.py index 15686662..a5733941 100644 --- a/parakeet/exps/ge2e/inference.py +++ b/paddlespeech/t2s/exps/ge2e/inference.py @@ -18,9 +18,9 @@ import numpy as np import paddle import tqdm -from parakeet.exps.ge2e.audio_processor import SpeakerVerificationPreprocessor -from parakeet.exps.ge2e.config import get_cfg_defaults -from parakeet.models.lstm_speaker_encoder import LSTMSpeakerEncoder +from paddlespeech.t2s.exps.ge2e.audio_processor import SpeakerVerificationPreprocessor +from paddlespeech.t2s.exps.ge2e.config import get_cfg_defaults +from paddlespeech.t2s.models.lstm_speaker_encoder import LSTMSpeakerEncoder def embed_utterance(processor, model, fpath_or_wav): diff --git a/parakeet/exps/ge2e/preprocess.py b/paddlespeech/t2s/exps/ge2e/preprocess.py similarity index 89% rename from parakeet/exps/ge2e/preprocess.py rename to paddlespeech/t2s/exps/ge2e/preprocess.py index f6457251..604ff0c6 100644 --- a/parakeet/exps/ge2e/preprocess.py +++ b/paddlespeech/t2s/exps/ge2e/preprocess.py @@ -16,12 +16,12 @@ from pathlib import Path from audio_processor import SpeakerVerificationPreprocessor -from parakeet.exps.ge2e.config import get_cfg_defaults -from parakeet.exps.ge2e.dataset_processors import process_aidatatang_200zh -from parakeet.exps.ge2e.dataset_processors import process_librispeech -from parakeet.exps.ge2e.dataset_processors import process_magicdata -from parakeet.exps.ge2e.dataset_processors import process_voxceleb1 -from parakeet.exps.ge2e.dataset_processors import process_voxceleb2 +from paddlespeech.t2s.exps.ge2e.config import get_cfg_defaults +from paddlespeech.t2s.exps.ge2e.dataset_processors import process_aidatatang_200zh +from paddlespeech.t2s.exps.ge2e.dataset_processors import process_librispeech +from paddlespeech.t2s.exps.ge2e.dataset_processors import process_magicdata +from paddlespeech.t2s.exps.ge2e.dataset_processors import process_voxceleb1 +from paddlespeech.t2s.exps.ge2e.dataset_processors import process_voxceleb2 if __name__ == "__main__": parser = argparse.ArgumentParser( diff --git a/parakeet/exps/ge2e/random_cycle.py b/paddlespeech/t2s/exps/ge2e/random_cycle.py similarity index 100% rename from parakeet/exps/ge2e/random_cycle.py rename to paddlespeech/t2s/exps/ge2e/random_cycle.py diff --git a/parakeet/exps/ge2e/speaker_verification_dataset.py b/paddlespeech/t2s/exps/ge2e/speaker_verification_dataset.py similarity index 98% rename from parakeet/exps/ge2e/speaker_verification_dataset.py rename to paddlespeech/t2s/exps/ge2e/speaker_verification_dataset.py index 896676d9..a1321996 100644 --- a/parakeet/exps/ge2e/speaker_verification_dataset.py +++ b/paddlespeech/t2s/exps/ge2e/speaker_verification_dataset.py @@ -18,7 +18,7 @@ import numpy as np from paddle.io import BatchSampler from paddle.io import Dataset -from parakeet.exps.ge2e.random_cycle import random_cycle +from paddlespeech.t2s.exps.ge2e.random_cycle import random_cycle class MultiSpeakerMelDataset(Dataset): diff --git a/parakeet/exps/ge2e/train.py b/paddlespeech/t2s/exps/ge2e/train.py similarity index 88% rename from parakeet/exps/ge2e/train.py rename to paddlespeech/t2s/exps/ge2e/train.py index 7a59c436..d3a57c93 100644 --- a/parakeet/exps/ge2e/train.py +++ b/paddlespeech/t2s/exps/ge2e/train.py @@ -19,13 +19,13 @@ from paddle.io import DataLoader from paddle.nn.clip import ClipGradByGlobalNorm from paddle.optimizer import Adam -from parakeet.exps.ge2e.config import get_cfg_defaults -from parakeet.exps.ge2e.speaker_verification_dataset import Collate -from parakeet.exps.ge2e.speaker_verification_dataset import MultiSpeakerMelDataset -from parakeet.exps.ge2e.speaker_verification_dataset import MultiSpeakerSampler -from parakeet.models.lstm_speaker_encoder import LSTMSpeakerEncoder -from parakeet.training import default_argument_parser -from parakeet.training import ExperimentBase +from paddlespeech.t2s.exps.ge2e.config import get_cfg_defaults +from paddlespeech.t2s.exps.ge2e.speaker_verification_dataset import Collate +from paddlespeech.t2s.exps.ge2e.speaker_verification_dataset import MultiSpeakerMelDataset +from paddlespeech.t2s.exps.ge2e.speaker_verification_dataset import MultiSpeakerSampler +from paddlespeech.t2s.models.lstm_speaker_encoder import LSTMSpeakerEncoder +from paddlespeech.t2s.training import default_argument_parser +from paddlespeech.t2s.training import ExperimentBase class Ge2eExperiment(ExperimentBase): diff --git a/parakeet/exps/sentences.txt b/paddlespeech/t2s/exps/sentences.txt similarity index 100% rename from parakeet/exps/sentences.txt rename to paddlespeech/t2s/exps/sentences.txt diff --git a/parakeet/exps/sentences_en.txt b/paddlespeech/t2s/exps/sentences_en.txt similarity index 100% rename from parakeet/exps/sentences_en.txt rename to paddlespeech/t2s/exps/sentences_en.txt diff --git a/parakeet/exps/speedyspeech/__init__.py b/paddlespeech/t2s/exps/speedyspeech/__init__.py similarity index 100% rename from parakeet/exps/speedyspeech/__init__.py rename to paddlespeech/t2s/exps/speedyspeech/__init__.py diff --git a/parakeet/exps/speedyspeech/inference.py b/paddlespeech/t2s/exps/speedyspeech/inference.py similarity index 98% rename from parakeet/exps/speedyspeech/inference.py rename to paddlespeech/t2s/exps/speedyspeech/inference.py index 77a90915..49ce37f2 100644 --- a/parakeet/exps/speedyspeech/inference.py +++ b/paddlespeech/t2s/exps/speedyspeech/inference.py @@ -18,7 +18,7 @@ from pathlib import Path import soundfile as sf from paddle import inference -from parakeet.frontend.zh_frontend import Frontend +from paddlespeech.t2s.frontend.zh_frontend import Frontend def main(): diff --git a/parakeet/exps/speedyspeech/normalize.py b/paddlespeech/t2s/exps/speedyspeech/normalize.py similarity index 98% rename from parakeet/exps/speedyspeech/normalize.py rename to paddlespeech/t2s/exps/speedyspeech/normalize.py index 8f02c33c..91d15c40 100644 --- a/parakeet/exps/speedyspeech/normalize.py +++ b/paddlespeech/t2s/exps/speedyspeech/normalize.py @@ -22,7 +22,7 @@ import numpy as np from sklearn.preprocessing import StandardScaler from tqdm import tqdm -from parakeet.datasets.data_table import DataTable +from paddlespeech.t2s.datasets.data_table import DataTable def main(): diff --git a/parakeet/exps/speedyspeech/preprocess.py b/paddlespeech/t2s/exps/speedyspeech/preprocess.py similarity index 96% rename from parakeet/exps/speedyspeech/preprocess.py rename to paddlespeech/t2s/exps/speedyspeech/preprocess.py index f3ae294d..aa589d5a 100644 --- a/parakeet/exps/speedyspeech/preprocess.py +++ b/paddlespeech/t2s/exps/speedyspeech/preprocess.py @@ -27,11 +27,11 @@ import tqdm import yaml from yacs.config import CfgNode -from parakeet.data.get_feats import LogMelFBank -from parakeet.datasets.preprocess_utils import compare_duration_and_mel_length -from parakeet.datasets.preprocess_utils import get_phn_dur -from parakeet.datasets.preprocess_utils import get_phones_tones -from parakeet.datasets.preprocess_utils import merge_silence +from paddlespeech.t2s.data.get_feats import LogMelFBank +from paddlespeech.t2s.datasets.preprocess_utils import compare_duration_and_mel_length +from paddlespeech.t2s.datasets.preprocess_utils import get_phn_dur +from paddlespeech.t2s.datasets.preprocess_utils import get_phones_tones +from paddlespeech.t2s.datasets.preprocess_utils import merge_silence def process_sentence(config: Dict[str, Any], diff --git a/parakeet/exps/speedyspeech/synthesize.py b/paddlespeech/t2s/exps/speedyspeech/synthesize.py similarity index 94% rename from parakeet/exps/speedyspeech/synthesize.py rename to paddlespeech/t2s/exps/speedyspeech/synthesize.py index 43ab4a69..4482c179 100644 --- a/parakeet/exps/speedyspeech/synthesize.py +++ b/paddlespeech/t2s/exps/speedyspeech/synthesize.py @@ -25,12 +25,12 @@ from paddle import jit from paddle.static import InputSpec from yacs.config import CfgNode -from parakeet.datasets.data_table import DataTable -from parakeet.models.parallel_wavegan import PWGGenerator -from parakeet.models.parallel_wavegan import PWGInference -from parakeet.models.speedyspeech import SpeedySpeech -from parakeet.models.speedyspeech import SpeedySpeechInference -from parakeet.modules.normalizer import ZScore +from paddlespeech.t2s.datasets.data_table import DataTable +from paddlespeech.t2s.models.parallel_wavegan import PWGGenerator +from paddlespeech.t2s.models.parallel_wavegan import PWGInference +from paddlespeech.t2s.models.speedyspeech import SpeedySpeech +from paddlespeech.t2s.models.speedyspeech import SpeedySpeechInference +from paddlespeech.t2s.modules.normalizer import ZScore def evaluate(args, speedyspeech_config, pwg_config): diff --git a/parakeet/exps/speedyspeech/synthesize_e2e.py b/paddlespeech/t2s/exps/speedyspeech/synthesize_e2e.py similarity index 94% rename from parakeet/exps/speedyspeech/synthesize_e2e.py rename to paddlespeech/t2s/exps/speedyspeech/synthesize_e2e.py index 47e064e9..0870d466 100644 --- a/parakeet/exps/speedyspeech/synthesize_e2e.py +++ b/paddlespeech/t2s/exps/speedyspeech/synthesize_e2e.py @@ -24,12 +24,12 @@ from paddle import jit from paddle.static import InputSpec from yacs.config import CfgNode -from parakeet.frontend.zh_frontend import Frontend -from parakeet.models.parallel_wavegan import PWGGenerator -from parakeet.models.parallel_wavegan import PWGInference -from parakeet.models.speedyspeech import SpeedySpeech -from parakeet.models.speedyspeech import SpeedySpeechInference -from parakeet.modules.normalizer import ZScore +from paddlespeech.t2s.frontend.zh_frontend import Frontend +from paddlespeech.t2s.models.parallel_wavegan import PWGGenerator +from paddlespeech.t2s.models.parallel_wavegan import PWGInference +from paddlespeech.t2s.models.speedyspeech import SpeedySpeech +from paddlespeech.t2s.models.speedyspeech import SpeedySpeechInference +from paddlespeech.t2s.modules.normalizer import ZScore def evaluate(args, speedyspeech_config, pwg_config): diff --git a/parakeet/exps/speedyspeech/train.py b/paddlespeech/t2s/exps/speedyspeech/train.py similarity index 91% rename from parakeet/exps/speedyspeech/train.py rename to paddlespeech/t2s/exps/speedyspeech/train.py index 6a4bf59e..772a39d7 100644 --- a/parakeet/exps/speedyspeech/train.py +++ b/paddlespeech/t2s/exps/speedyspeech/train.py @@ -27,16 +27,16 @@ from paddle.io import DataLoader from paddle.io import DistributedBatchSampler from yacs.config import CfgNode -from parakeet.datasets.am_batch_fn import speedyspeech_batch_fn -from parakeet.datasets.data_table import DataTable -from parakeet.models.speedyspeech import SpeedySpeech -from parakeet.models.speedyspeech import SpeedySpeechEvaluator -from parakeet.models.speedyspeech import SpeedySpeechUpdater -from parakeet.training.extensions.snapshot import Snapshot -from parakeet.training.extensions.visualizer import VisualDL -from parakeet.training.optimizer import build_optimizers -from parakeet.training.seeding import seed_everything -from parakeet.training.trainer import Trainer +from paddlespeech.t2s.datasets.am_batch_fn import speedyspeech_batch_fn +from paddlespeech.t2s.datasets.data_table import DataTable +from paddlespeech.t2s.models.speedyspeech import SpeedySpeech +from paddlespeech.t2s.models.speedyspeech import SpeedySpeechEvaluator +from paddlespeech.t2s.models.speedyspeech import SpeedySpeechUpdater +from paddlespeech.t2s.training.extensions.snapshot import Snapshot +from paddlespeech.t2s.training.extensions.visualizer import VisualDL +from paddlespeech.t2s.training.optimizer import build_optimizers +from paddlespeech.t2s.training.seeding import seed_everything +from paddlespeech.t2s.training.trainer import Trainer def train_sp(args, config): diff --git a/parakeet/exps/tacotron2/__init__.py b/paddlespeech/t2s/exps/tacotron2/__init__.py similarity index 100% rename from parakeet/exps/tacotron2/__init__.py rename to paddlespeech/t2s/exps/tacotron2/__init__.py diff --git a/parakeet/exps/tacotron2/config.py b/paddlespeech/t2s/exps/tacotron2/config.py similarity index 100% rename from parakeet/exps/tacotron2/config.py rename to paddlespeech/t2s/exps/tacotron2/config.py diff --git a/parakeet/exps/tacotron2/ljspeech.py b/paddlespeech/t2s/exps/tacotron2/ljspeech.py similarity index 96% rename from parakeet/exps/tacotron2/ljspeech.py rename to paddlespeech/t2s/exps/tacotron2/ljspeech.py index 59c855eb..08db2a64 100644 --- a/parakeet/exps/tacotron2/ljspeech.py +++ b/paddlespeech/t2s/exps/tacotron2/ljspeech.py @@ -17,8 +17,8 @@ from pathlib import Path import numpy as np from paddle.io import Dataset -from parakeet.data.batch import batch_spec -from parakeet.data.batch import batch_text_id +from paddlespeech.t2s.data.batch import batch_spec +from paddlespeech.t2s.data.batch import batch_text_id class LJSpeech(Dataset): diff --git a/parakeet/exps/tacotron2/preprocess.py b/paddlespeech/t2s/exps/tacotron2/preprocess.py similarity index 91% rename from parakeet/exps/tacotron2/preprocess.py rename to paddlespeech/t2s/exps/tacotron2/preprocess.py index 89344485..480b3331 100644 --- a/parakeet/exps/tacotron2/preprocess.py +++ b/paddlespeech/t2s/exps/tacotron2/preprocess.py @@ -19,11 +19,11 @@ from pathlib import Path import numpy as np import tqdm -from parakeet.audio import AudioProcessor -from parakeet.audio import LogMagnitude -from parakeet.datasets import LJSpeechMetaData -from parakeet.exps.tacotron2.config import get_cfg_defaults -from parakeet.frontend import EnglishCharacter +from paddlespeech.t2s.audio import AudioProcessor +from paddlespeech.t2s.audio import LogMagnitude +from paddlespeech.t2s.datasets import LJSpeechMetaData +from paddlespeech.t2s.exps.tacotron2.config import get_cfg_defaults +from paddlespeech.t2s.frontend import EnglishCharacter def create_dataset(config, source_path, target_path, verbose=False): diff --git a/parakeet/exps/tacotron2/synthesize.ipynb b/paddlespeech/t2s/exps/tacotron2/synthesize.ipynb similarity index 99% rename from parakeet/exps/tacotron2/synthesize.ipynb rename to paddlespeech/t2s/exps/tacotron2/synthesize.ipynb index f71182f2..cc424311 100644 --- a/parakeet/exps/tacotron2/synthesize.ipynb +++ b/paddlespeech/t2s/exps/tacotron2/synthesize.ipynb @@ -19,8 +19,8 @@ "from IPython import display as ipd\n", "%matplotlib inline\n", "\n", - "from parakeet.utils import display\n", - "from parakeet.utils import layer_tools\n", + "from paddlespeech.t2s.utils import display\n", + "from paddlespeech.t2s.utils import layer_tools\n", "paddle.set_device(\"gpu:0\")\n", "\n", "import sys\n", @@ -45,8 +45,8 @@ "metadata": {}, "outputs": [], "source": [ - "from parakeet.models.tacotron2 import Tacotron2\n", - "from parakeet.frontend import EnglishCharacter" + "from paddlespeech.t2s.models.tacotron2 import Tacotron2\n", + "from paddlespeech.t2s.frontend import EnglishCharacter" ] }, { @@ -198,7 +198,7 @@ "metadata": {}, "outputs": [], "source": [ - "from parakeet.models.waveflow import ConditionalWaveFlow" + "from paddlespeech.t2s.models.waveflow import ConditionalWaveFlow" ] }, { diff --git a/parakeet/exps/tacotron2/synthesize.py b/paddlespeech/t2s/exps/tacotron2/synthesize.py similarity index 93% rename from parakeet/exps/tacotron2/synthesize.py rename to paddlespeech/t2s/exps/tacotron2/synthesize.py index 56257c9b..613fec02 100644 --- a/parakeet/exps/tacotron2/synthesize.py +++ b/paddlespeech/t2s/exps/tacotron2/synthesize.py @@ -18,10 +18,10 @@ import numpy as np import paddle from matplotlib import pyplot as plt -from parakeet.exps.tacotron2.config import get_cfg_defaults -from parakeet.frontend import EnglishCharacter -from parakeet.models.tacotron2 import Tacotron2 -from parakeet.utils import display +from paddlespeech.t2s.exps.tacotron2.config import get_cfg_defaults +from paddlespeech.t2s.frontend import EnglishCharacter +from paddlespeech.t2s.models.tacotron2 import Tacotron2 +from paddlespeech.t2s.utils import display def main(config, args): diff --git a/parakeet/exps/tacotron2/train.py b/paddlespeech/t2s/exps/tacotron2/train.py similarity index 93% rename from parakeet/exps/tacotron2/train.py rename to paddlespeech/t2s/exps/tacotron2/train.py index 3677c271..a5f08360 100644 --- a/parakeet/exps/tacotron2/train.py +++ b/paddlespeech/t2s/exps/tacotron2/train.py @@ -20,16 +20,16 @@ from paddle import distributed as dist from paddle.io import DataLoader from paddle.io import DistributedBatchSampler -from parakeet.data import dataset -from parakeet.exps.tacotron2.config import get_cfg_defaults -from parakeet.exps.tacotron2.ljspeech import LJSpeech -from parakeet.exps.tacotron2.ljspeech import LJSpeechCollector -from parakeet.models.tacotron2 import Tacotron2 -from parakeet.models.tacotron2 import Tacotron2Loss -from parakeet.training.cli import default_argument_parser -from parakeet.training.experiment import ExperimentBase -from parakeet.utils import display -from parakeet.utils import mp_tools +from paddlespeech.t2s.data import dataset +from paddlespeech.t2s.exps.tacotron2.config import get_cfg_defaults +from paddlespeech.t2s.exps.tacotron2.ljspeech import LJSpeech +from paddlespeech.t2s.exps.tacotron2.ljspeech import LJSpeechCollector +from paddlespeech.t2s.models.tacotron2 import Tacotron2 +from paddlespeech.t2s.models.tacotron2 import Tacotron2Loss +from paddlespeech.t2s.training.cli import default_argument_parser +from paddlespeech.t2s.training.experiment import ExperimentBase +from paddlespeech.t2s.utils import display +from paddlespeech.t2s.utils import mp_tools class Experiment(ExperimentBase): diff --git a/parakeet/exps/transformer_tts/__init__.py b/paddlespeech/t2s/exps/transformer_tts/__init__.py similarity index 100% rename from parakeet/exps/transformer_tts/__init__.py rename to paddlespeech/t2s/exps/transformer_tts/__init__.py diff --git a/parakeet/exps/transformer_tts/normalize.py b/paddlespeech/t2s/exps/transformer_tts/normalize.py similarity index 98% rename from parakeet/exps/transformer_tts/normalize.py rename to paddlespeech/t2s/exps/transformer_tts/normalize.py index 127449ee..4bb77c79 100644 --- a/parakeet/exps/transformer_tts/normalize.py +++ b/paddlespeech/t2s/exps/transformer_tts/normalize.py @@ -22,7 +22,7 @@ import numpy as np from sklearn.preprocessing import StandardScaler from tqdm import tqdm -from parakeet.datasets.data_table import DataTable +from paddlespeech.t2s.datasets.data_table import DataTable def main(): diff --git a/parakeet/exps/transformer_tts/preprocess.py b/paddlespeech/t2s/exps/transformer_tts/preprocess.py similarity index 98% rename from parakeet/exps/transformer_tts/preprocess.py rename to paddlespeech/t2s/exps/transformer_tts/preprocess.py index 96696eae..93158b67 100644 --- a/parakeet/exps/transformer_tts/preprocess.py +++ b/paddlespeech/t2s/exps/transformer_tts/preprocess.py @@ -26,8 +26,8 @@ import tqdm import yaml from yacs.config import CfgNode as Configuration -from parakeet.data.get_feats import LogMelFBank -from parakeet.frontend import English +from paddlespeech.t2s.data.get_feats import LogMelFBank +from paddlespeech.t2s.frontend import English def get_lj_sentences(file_name, frontend): diff --git a/parakeet/exps/transformer_tts/synthesize.py b/paddlespeech/t2s/exps/transformer_tts/synthesize.py similarity index 92% rename from parakeet/exps/transformer_tts/synthesize.py rename to paddlespeech/t2s/exps/transformer_tts/synthesize.py index 5c1945d2..82fd8f15 100644 --- a/parakeet/exps/transformer_tts/synthesize.py +++ b/paddlespeech/t2s/exps/transformer_tts/synthesize.py @@ -22,12 +22,12 @@ import soundfile as sf import yaml from yacs.config import CfgNode -from parakeet.datasets.data_table import DataTable -from parakeet.models.transformer_tts import TransformerTTS -from parakeet.models.transformer_tts import TransformerTTSInference -from parakeet.models.waveflow import ConditionalWaveFlow -from parakeet.modules.normalizer import ZScore -from parakeet.utils import layer_tools +from paddlespeech.t2s.datasets.data_table import DataTable +from paddlespeech.t2s.models.transformer_tts import TransformerTTS +from paddlespeech.t2s.models.transformer_tts import TransformerTTSInference +from paddlespeech.t2s.models.waveflow import ConditionalWaveFlow +from paddlespeech.t2s.modules.normalizer import ZScore +from paddlespeech.t2s.utils import layer_tools def evaluate(args, acoustic_model_config, vocoder_config): diff --git a/parakeet/exps/transformer_tts/synthesize_e2e.py b/paddlespeech/t2s/exps/transformer_tts/synthesize_e2e.py similarity index 93% rename from parakeet/exps/transformer_tts/synthesize_e2e.py rename to paddlespeech/t2s/exps/transformer_tts/synthesize_e2e.py index 2bee77d3..993749f0 100644 --- a/parakeet/exps/transformer_tts/synthesize_e2e.py +++ b/paddlespeech/t2s/exps/transformer_tts/synthesize_e2e.py @@ -21,12 +21,12 @@ import soundfile as sf import yaml from yacs.config import CfgNode -from parakeet.frontend import English -from parakeet.models.transformer_tts import TransformerTTS -from parakeet.models.transformer_tts import TransformerTTSInference -from parakeet.models.waveflow import ConditionalWaveFlow -from parakeet.modules.normalizer import ZScore -from parakeet.utils import layer_tools +from paddlespeech.t2s.frontend import English +from paddlespeech.t2s.models.transformer_tts import TransformerTTS +from paddlespeech.t2s.models.transformer_tts import TransformerTTSInference +from paddlespeech.t2s.models.waveflow import ConditionalWaveFlow +from paddlespeech.t2s.modules.normalizer import ZScore +from paddlespeech.t2s.utils import layer_tools def evaluate(args, acoustic_model_config, vocoder_config): diff --git a/parakeet/exps/transformer_tts/train.py b/paddlespeech/t2s/exps/transformer_tts/train.py similarity index 89% rename from parakeet/exps/transformer_tts/train.py rename to paddlespeech/t2s/exps/transformer_tts/train.py index bf066390..7d9020a3 100644 --- a/parakeet/exps/transformer_tts/train.py +++ b/paddlespeech/t2s/exps/transformer_tts/train.py @@ -27,16 +27,16 @@ from paddle.io import DataLoader from paddle.io import DistributedBatchSampler from yacs.config import CfgNode -from parakeet.datasets.am_batch_fn import transformer_single_spk_batch_fn -from parakeet.datasets.data_table import DataTable -from parakeet.models.transformer_tts import TransformerTTS -from parakeet.models.transformer_tts import TransformerTTSEvaluator -from parakeet.models.transformer_tts import TransformerTTSUpdater -from parakeet.training.extensions.snapshot import Snapshot -from parakeet.training.extensions.visualizer import VisualDL -from parakeet.training.optimizer import build_optimizers -from parakeet.training.seeding import seed_everything -from parakeet.training.trainer import Trainer +from paddlespeech.t2s.datasets.am_batch_fn import transformer_single_spk_batch_fn +from paddlespeech.t2s.datasets.data_table import DataTable +from paddlespeech.t2s.models.transformer_tts import TransformerTTS +from paddlespeech.t2s.models.transformer_tts import TransformerTTSEvaluator +from paddlespeech.t2s.models.transformer_tts import TransformerTTSUpdater +from paddlespeech.t2s.training.extensions.snapshot import Snapshot +from paddlespeech.t2s.training.extensions.visualizer import VisualDL +from paddlespeech.t2s.training.optimizer import build_optimizers +from paddlespeech.t2s.training.seeding import seed_everything +from paddlespeech.t2s.training.trainer import Trainer def train_sp(args, config): diff --git a/parakeet/exps/voice_cloning/__init__.py b/paddlespeech/t2s/exps/voice_cloning/__init__.py similarity index 100% rename from parakeet/exps/voice_cloning/__init__.py rename to paddlespeech/t2s/exps/voice_cloning/__init__.py diff --git a/parakeet/exps/voice_cloning/tacotron2_ge2e/__init__.py b/paddlespeech/t2s/exps/voice_cloning/tacotron2_ge2e/__init__.py similarity index 100% rename from parakeet/exps/voice_cloning/tacotron2_ge2e/__init__.py rename to paddlespeech/t2s/exps/voice_cloning/tacotron2_ge2e/__init__.py diff --git a/parakeet/exps/voice_cloning/tacotron2_ge2e/aishell3.py b/paddlespeech/t2s/exps/voice_cloning/tacotron2_ge2e/aishell3.py similarity index 89% rename from parakeet/exps/voice_cloning/tacotron2_ge2e/aishell3.py rename to paddlespeech/t2s/exps/voice_cloning/tacotron2_ge2e/aishell3.py index b697ecf3..da95582d 100644 --- a/parakeet/exps/voice_cloning/tacotron2_ge2e/aishell3.py +++ b/paddlespeech/t2s/exps/voice_cloning/tacotron2_ge2e/aishell3.py @@ -17,11 +17,11 @@ from pathlib import Path import numpy as np from paddle.io import Dataset -from parakeet.data import batch_spec -from parakeet.data import batch_text_id -from parakeet.exps.voice_cloning.tacotron2_ge2e.preprocess_transcription import _phones -from parakeet.exps.voice_cloning.tacotron2_ge2e.preprocess_transcription import _tones -from parakeet.frontend import Vocab +from paddlespeech.t2s.data import batch_spec +from paddlespeech.t2s.data import batch_text_id +from paddlespeech.t2s.exps.voice_cloning.tacotron2_ge2e.preprocess_transcription import _phones +from paddlespeech.t2s.exps.voice_cloning.tacotron2_ge2e.preprocess_transcription import _tones +from paddlespeech.t2s.frontend import Vocab voc_phones = Vocab(sorted(list(_phones))) print("vocab_phones:\n", voc_phones) diff --git a/parakeet/exps/voice_cloning/tacotron2_ge2e/chinese_g2p.py b/paddlespeech/t2s/exps/voice_cloning/tacotron2_ge2e/chinese_g2p.py similarity index 92% rename from parakeet/exps/voice_cloning/tacotron2_ge2e/chinese_g2p.py rename to paddlespeech/t2s/exps/voice_cloning/tacotron2_ge2e/chinese_g2p.py index f975522f..12de3bb7 100644 --- a/parakeet/exps/voice_cloning/tacotron2_ge2e/chinese_g2p.py +++ b/paddlespeech/t2s/exps/voice_cloning/tacotron2_ge2e/chinese_g2p.py @@ -17,7 +17,7 @@ from typing import Tuple from pypinyin import lazy_pinyin from pypinyin import Style -from parakeet.exps.voice_cloning.tacotron2_ge2e.preprocess_transcription import split_syllable +from paddlespeech.t2s.exps.voice_cloning.tacotron2_ge2e.preprocess_transcription import split_syllable def convert_to_pinyin(text: str) -> List[str]: diff --git a/parakeet/exps/voice_cloning/tacotron2_ge2e/config.py b/paddlespeech/t2s/exps/voice_cloning/tacotron2_ge2e/config.py similarity index 100% rename from parakeet/exps/voice_cloning/tacotron2_ge2e/config.py rename to paddlespeech/t2s/exps/voice_cloning/tacotron2_ge2e/config.py diff --git a/parakeet/exps/voice_cloning/tacotron2_ge2e/extract_mel.py b/paddlespeech/t2s/exps/voice_cloning/tacotron2_ge2e/extract_mel.py similarity index 92% rename from parakeet/exps/voice_cloning/tacotron2_ge2e/extract_mel.py rename to paddlespeech/t2s/exps/voice_cloning/tacotron2_ge2e/extract_mel.py index 53daa3fc..d12466f6 100644 --- a/parakeet/exps/voice_cloning/tacotron2_ge2e/extract_mel.py +++ b/paddlespeech/t2s/exps/voice_cloning/tacotron2_ge2e/extract_mel.py @@ -19,10 +19,10 @@ from pathlib import Path import numpy as np import tqdm -from parakeet.audio import AudioProcessor -from parakeet.audio.spec_normalizer import LogMagnitude -from parakeet.audio.spec_normalizer import NormalizerBase -from parakeet.exps.voice_cloning.tacotron2_ge2e.config import get_cfg_defaults +from paddlespeech.t2s.audio import AudioProcessor +from paddlespeech.t2s.audio.spec_normalizer import LogMagnitude +from paddlespeech.t2s.audio.spec_normalizer import NormalizerBase +from paddlespeech.t2s.exps.voice_cloning.tacotron2_ge2e.config import get_cfg_defaults def extract_mel(fname: Path, diff --git a/parakeet/exps/voice_cloning/tacotron2_ge2e/lexicon.txt b/paddlespeech/t2s/exps/voice_cloning/tacotron2_ge2e/lexicon.txt similarity index 100% rename from parakeet/exps/voice_cloning/tacotron2_ge2e/lexicon.txt rename to paddlespeech/t2s/exps/voice_cloning/tacotron2_ge2e/lexicon.txt diff --git a/parakeet/exps/voice_cloning/tacotron2_ge2e/preprocess_transcription.py b/paddlespeech/t2s/exps/voice_cloning/tacotron2_ge2e/preprocess_transcription.py similarity index 100% rename from parakeet/exps/voice_cloning/tacotron2_ge2e/preprocess_transcription.py rename to paddlespeech/t2s/exps/voice_cloning/tacotron2_ge2e/preprocess_transcription.py diff --git a/parakeet/exps/voice_cloning/tacotron2_ge2e/process_wav.py b/paddlespeech/t2s/exps/voice_cloning/tacotron2_ge2e/process_wav.py similarity index 100% rename from parakeet/exps/voice_cloning/tacotron2_ge2e/process_wav.py rename to paddlespeech/t2s/exps/voice_cloning/tacotron2_ge2e/process_wav.py diff --git a/parakeet/exps/voice_cloning/tacotron2_ge2e/train.py b/paddlespeech/t2s/exps/voice_cloning/tacotron2_ge2e/train.py similarity index 93% rename from parakeet/exps/voice_cloning/tacotron2_ge2e/train.py rename to paddlespeech/t2s/exps/voice_cloning/tacotron2_ge2e/train.py index 1a9bd8cb..ceae1360 100644 --- a/parakeet/exps/voice_cloning/tacotron2_ge2e/train.py +++ b/paddlespeech/t2s/exps/voice_cloning/tacotron2_ge2e/train.py @@ -22,16 +22,16 @@ from paddle import distributed as dist from paddle.io import DataLoader from paddle.io import DistributedBatchSampler -from parakeet.data import dataset -from parakeet.exps.voice_cloning.tacotron2_ge2e.aishell3 import AiShell3 -from parakeet.exps.voice_cloning.tacotron2_ge2e.aishell3 import collate_aishell3_examples -from parakeet.exps.voice_cloning.tacotron2_ge2e.config import get_cfg_defaults -from parakeet.models.tacotron2 import Tacotron2 -from parakeet.models.tacotron2 import Tacotron2Loss -from parakeet.training.cli import default_argument_parser -from parakeet.training.experiment import ExperimentBase -from parakeet.utils import display -from parakeet.utils import mp_tools +from paddlespeech.t2s.data import dataset +from paddlespeech.t2s.exps.voice_cloning.tacotron2_ge2e.aishell3 import AiShell3 +from paddlespeech.t2s.exps.voice_cloning.tacotron2_ge2e.aishell3 import collate_aishell3_examples +from paddlespeech.t2s.exps.voice_cloning.tacotron2_ge2e.config import get_cfg_defaults +from paddlespeech.t2s.models.tacotron2 import Tacotron2 +from paddlespeech.t2s.models.tacotron2 import Tacotron2Loss +from paddlespeech.t2s.training.cli import default_argument_parser +from paddlespeech.t2s.training.experiment import ExperimentBase +from paddlespeech.t2s.utils import display +from paddlespeech.t2s.utils import mp_tools class Experiment(ExperimentBase): diff --git a/parakeet/exps/voice_cloning/tacotron2_ge2e/voice_cloning.py b/paddlespeech/t2s/exps/voice_cloning/tacotron2_ge2e/voice_cloning.py similarity index 90% rename from parakeet/exps/voice_cloning/tacotron2_ge2e/voice_cloning.py rename to paddlespeech/t2s/exps/voice_cloning/tacotron2_ge2e/voice_cloning.py index 8afd35b7..c76ce007 100644 --- a/parakeet/exps/voice_cloning/tacotron2_ge2e/voice_cloning.py +++ b/paddlespeech/t2s/exps/voice_cloning/tacotron2_ge2e/voice_cloning.py @@ -20,14 +20,14 @@ import paddle import soundfile as sf from matplotlib import pyplot as plt -from parakeet.exps.ge2e.audio_processor import SpeakerVerificationPreprocessor -from parakeet.exps.voice_cloning.tacotron2_ge2e.aishell3 import voc_phones -from parakeet.exps.voice_cloning.tacotron2_ge2e.aishell3 import voc_tones -from parakeet.exps.voice_cloning.tacotron2_ge2e.chinese_g2p import convert_sentence -from parakeet.models.lstm_speaker_encoder import LSTMSpeakerEncoder -from parakeet.models.tacotron2 import Tacotron2 -from parakeet.models.waveflow import ConditionalWaveFlow -from parakeet.utils import display +from paddlespeech.t2s.exps.ge2e.audio_processor import SpeakerVerificationPreprocessor +from paddlespeech.t2s.exps.voice_cloning.tacotron2_ge2e.aishell3 import voc_phones +from paddlespeech.t2s.exps.voice_cloning.tacotron2_ge2e.aishell3 import voc_tones +from paddlespeech.t2s.exps.voice_cloning.tacotron2_ge2e.chinese_g2p import convert_sentence +from paddlespeech.t2s.models.lstm_speaker_encoder import LSTMSpeakerEncoder +from paddlespeech.t2s.models.tacotron2 import Tacotron2 +from paddlespeech.t2s.models.waveflow import ConditionalWaveFlow +from paddlespeech.t2s.utils import display def voice_cloning(args): diff --git a/parakeet/exps/waveflow/__init__.py b/paddlespeech/t2s/exps/waveflow/__init__.py similarity index 100% rename from parakeet/exps/waveflow/__init__.py rename to paddlespeech/t2s/exps/waveflow/__init__.py diff --git a/parakeet/exps/waveflow/config.py b/paddlespeech/t2s/exps/waveflow/config.py similarity index 100% rename from parakeet/exps/waveflow/config.py rename to paddlespeech/t2s/exps/waveflow/config.py diff --git a/parakeet/exps/waveflow/ljspeech.py b/paddlespeech/t2s/exps/waveflow/ljspeech.py similarity index 96% rename from parakeet/exps/waveflow/ljspeech.py rename to paddlespeech/t2s/exps/waveflow/ljspeech.py index ca18f400..655b63da 100644 --- a/parakeet/exps/waveflow/ljspeech.py +++ b/paddlespeech/t2s/exps/waveflow/ljspeech.py @@ -17,8 +17,8 @@ import numpy as np import pandas from paddle.io import Dataset -from parakeet.data.batch import batch_spec -from parakeet.data.batch import batch_wav +from paddlespeech.t2s.data.batch import batch_spec +from paddlespeech.t2s.data.batch import batch_wav class LJSpeech(Dataset): diff --git a/parakeet/exps/waveflow/preprocess.py b/paddlespeech/t2s/exps/waveflow/preprocess.py similarity index 96% rename from parakeet/exps/waveflow/preprocess.py rename to paddlespeech/t2s/exps/waveflow/preprocess.py index d4ec0de5..ef3a2917 100644 --- a/parakeet/exps/waveflow/preprocess.py +++ b/paddlespeech/t2s/exps/waveflow/preprocess.py @@ -20,9 +20,9 @@ import numpy as np import pandas as pd import tqdm -from parakeet.audio import LogMagnitude -from parakeet.datasets import LJSpeechMetaData -from parakeet.exps.waveflow.config import get_cfg_defaults +from paddlespeech.t2s.audio import LogMagnitude +from paddlespeech.t2s.datasets import LJSpeechMetaData +from paddlespeech.t2s.exps.waveflow.config import get_cfg_defaults class Transform(object): diff --git a/parakeet/exps/waveflow/synthesize.py b/paddlespeech/t2s/exps/waveflow/synthesize.py similarity index 93% rename from parakeet/exps/waveflow/synthesize.py rename to paddlespeech/t2s/exps/waveflow/synthesize.py index 4dd52514..4f07aa4e 100644 --- a/parakeet/exps/waveflow/synthesize.py +++ b/paddlespeech/t2s/exps/waveflow/synthesize.py @@ -19,9 +19,9 @@ import numpy as np import paddle import soundfile as sf -from parakeet.exps.waveflow.config import get_cfg_defaults -from parakeet.models.waveflow import ConditionalWaveFlow -from parakeet.utils import layer_tools +from paddlespeech.t2s.exps.waveflow.config import get_cfg_defaults +from paddlespeech.t2s.models.waveflow import ConditionalWaveFlow +from paddlespeech.t2s.utils import layer_tools def main(config, args): diff --git a/parakeet/exps/waveflow/train.py b/paddlespeech/t2s/exps/waveflow/train.py similarity index 89% rename from parakeet/exps/waveflow/train.py rename to paddlespeech/t2s/exps/waveflow/train.py index ecfcbcaa..9d1df13c 100644 --- a/parakeet/exps/waveflow/train.py +++ b/paddlespeech/t2s/exps/waveflow/train.py @@ -19,16 +19,16 @@ from paddle import distributed as dist from paddle.io import DataLoader from paddle.io import DistributedBatchSampler -from parakeet.data import dataset -from parakeet.exps.waveflow.config import get_cfg_defaults -from parakeet.exps.waveflow.ljspeech import LJSpeech -from parakeet.exps.waveflow.ljspeech import LJSpeechClipCollector -from parakeet.exps.waveflow.ljspeech import LJSpeechCollector -from parakeet.models.waveflow import ConditionalWaveFlow -from parakeet.models.waveflow import WaveFlowLoss -from parakeet.training.cli import default_argument_parser -from parakeet.training.experiment import ExperimentBase -from parakeet.utils import mp_tools +from paddlespeech.t2s.data import dataset +from paddlespeech.t2s.exps.waveflow.config import get_cfg_defaults +from paddlespeech.t2s.exps.waveflow.ljspeech import LJSpeech +from paddlespeech.t2s.exps.waveflow.ljspeech import LJSpeechClipCollector +from paddlespeech.t2s.exps.waveflow.ljspeech import LJSpeechCollector +from paddlespeech.t2s.models.waveflow import ConditionalWaveFlow +from paddlespeech.t2s.models.waveflow import WaveFlowLoss +from paddlespeech.t2s.training.cli import default_argument_parser +from paddlespeech.t2s.training.experiment import ExperimentBase +from paddlespeech.t2s.utils import mp_tools class Experiment(ExperimentBase): diff --git a/parakeet/frontend/__init__.py b/paddlespeech/t2s/frontend/__init__.py similarity index 100% rename from parakeet/frontend/__init__.py rename to paddlespeech/t2s/frontend/__init__.py diff --git a/parakeet/frontend/arpabet.py b/paddlespeech/t2s/frontend/arpabet.py similarity index 98% rename from parakeet/frontend/arpabet.py rename to paddlespeech/t2s/frontend/arpabet.py index ae9212b2..094a2bfa 100644 --- a/parakeet/frontend/arpabet.py +++ b/paddlespeech/t2s/frontend/arpabet.py @@ -11,7 +11,7 @@ # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. -from parakeet.frontend.phonectic import Phonetics +from paddlespeech.t2s.frontend.phonectic import Phonetics """ A phonology system with ARPABET symbols and limited punctuations. The G2P conversion is done by g2p_en. @@ -19,7 +19,7 @@ conversion is done by g2p_en. Note that g2p_en does not handle words with hypen well. So make sure the input sentence is first normalized. """ -from parakeet.frontend.vocab import Vocab +from paddlespeech.t2s.frontend.vocab import Vocab from g2p_en import G2p diff --git a/parakeet/frontend/generate_lexicon.py b/paddlespeech/t2s/frontend/generate_lexicon.py similarity index 100% rename from parakeet/frontend/generate_lexicon.py rename to paddlespeech/t2s/frontend/generate_lexicon.py diff --git a/paddlespeech/t2s/frontend/normalizer/__init__.py b/paddlespeech/t2s/frontend/normalizer/__init__.py new file mode 100644 index 00000000..a03329f1 --- /dev/null +++ b/paddlespeech/t2s/frontend/normalizer/__init__.py @@ -0,0 +1,15 @@ +# Copyright (c) 2020 PaddlePaddle Authors. All Rights Reserved. +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +from paddlespeech.t2s.frontend.normalizer.normalizer import * +from paddlespeech.t2s.frontend.normalizer.numbers import * diff --git a/parakeet/frontend/normalizer/abbrrviation.py b/paddlespeech/t2s/frontend/normalizer/abbrrviation.py similarity index 100% rename from parakeet/frontend/normalizer/abbrrviation.py rename to paddlespeech/t2s/frontend/normalizer/abbrrviation.py diff --git a/parakeet/frontend/normalizer/acronyms.py b/paddlespeech/t2s/frontend/normalizer/acronyms.py similarity index 100% rename from parakeet/frontend/normalizer/acronyms.py rename to paddlespeech/t2s/frontend/normalizer/acronyms.py diff --git a/parakeet/frontend/normalizer/normalizer.py b/paddlespeech/t2s/frontend/normalizer/normalizer.py similarity index 94% rename from parakeet/frontend/normalizer/normalizer.py rename to paddlespeech/t2s/frontend/normalizer/normalizer.py index 795607e5..421ebd1f 100644 --- a/parakeet/frontend/normalizer/normalizer.py +++ b/paddlespeech/t2s/frontend/normalizer/normalizer.py @@ -15,7 +15,7 @@ import re import unicodedata from builtins import str as unicode -from parakeet.frontend.normalizer.numbers import normalize_numbers +from paddlespeech.t2s.frontend.normalizer.numbers import normalize_numbers def normalize(sentence): diff --git a/parakeet/frontend/normalizer/numbers.py b/paddlespeech/t2s/frontend/normalizer/numbers.py similarity index 100% rename from parakeet/frontend/normalizer/numbers.py rename to paddlespeech/t2s/frontend/normalizer/numbers.py diff --git a/parakeet/frontend/normalizer/width.py b/paddlespeech/t2s/frontend/normalizer/width.py similarity index 100% rename from parakeet/frontend/normalizer/width.py rename to paddlespeech/t2s/frontend/normalizer/width.py diff --git a/parakeet/frontend/phonectic.py b/paddlespeech/t2s/frontend/phonectic.py similarity index 97% rename from parakeet/frontend/phonectic.py rename to paddlespeech/t2s/frontend/phonectic.py index 874c1979..8eac0b48 100644 --- a/parakeet/frontend/phonectic.py +++ b/paddlespeech/t2s/frontend/phonectic.py @@ -17,9 +17,9 @@ from abc import abstractmethod from g2p_en import G2p from g2pM import G2pM -from parakeet.frontend.normalizer.normalizer import normalize -from parakeet.frontend.punctuation import get_punctuations -from parakeet.frontend.vocab import Vocab +from paddlespeech.t2s.frontend.normalizer.normalizer import normalize +from paddlespeech.t2s.frontend.punctuation import get_punctuations +from paddlespeech.t2s.frontend.vocab import Vocab # discard opencc untill we find an easy solution to install it on windows # from opencc import OpenCC diff --git a/parakeet/frontend/pinyin.py b/paddlespeech/t2s/frontend/pinyin.py similarity index 98% rename from parakeet/frontend/pinyin.py rename to paddlespeech/t2s/frontend/pinyin.py index 503bfd5d..f99129ce 100644 --- a/parakeet/frontend/pinyin.py +++ b/paddlespeech/t2s/frontend/pinyin.py @@ -26,8 +26,8 @@ from pypinyin.core import DefaultConverter from pypinyin.core import Pinyin from pypinyin.core import Style -from parakeet.frontend.phonectic import Phonetics -from parakeet.frontend.vocab import Vocab +from paddlespeech.t2s.frontend.phonectic import Phonetics +from paddlespeech.t2s.frontend.vocab import Vocab _punctuations = ['，', '。', '？', '！'] _initials = [ diff --git a/parakeet/frontend/punctuation.py b/paddlespeech/t2s/frontend/punctuation.py similarity index 100% rename from parakeet/frontend/punctuation.py rename to paddlespeech/t2s/frontend/punctuation.py diff --git a/parakeet/frontend/tone_sandhi.py b/paddlespeech/t2s/frontend/tone_sandhi.py similarity index 100% rename from parakeet/frontend/tone_sandhi.py rename to paddlespeech/t2s/frontend/tone_sandhi.py diff --git a/parakeet/frontend/vocab.py b/paddlespeech/t2s/frontend/vocab.py similarity index 100% rename from parakeet/frontend/vocab.py rename to paddlespeech/t2s/frontend/vocab.py diff --git a/parakeet/frontend/zh_frontend.py b/paddlespeech/t2s/frontend/zh_frontend.py similarity index 98% rename from parakeet/frontend/zh_frontend.py rename to paddlespeech/t2s/frontend/zh_frontend.py index 04ce235f..33bf5ab2 100644 --- a/parakeet/frontend/zh_frontend.py +++ b/paddlespeech/t2s/frontend/zh_frontend.py @@ -22,9 +22,9 @@ from g2pM import G2pM from pypinyin import lazy_pinyin from pypinyin import Style -from parakeet.frontend.generate_lexicon import generate_lexicon -from parakeet.frontend.tone_sandhi import ToneSandhi -from parakeet.frontend.zh_normalization.text_normlization import TextNormalizer +from paddlespeech.t2s.frontend.generate_lexicon import generate_lexicon +from paddlespeech.t2s.frontend.tone_sandhi import ToneSandhi +from paddlespeech.t2s.frontend.zh_normalization.text_normlization import TextNormalizer class Frontend(): diff --git a/parakeet/frontend/zh_normalization/README.md b/paddlespeech/t2s/frontend/zh_normalization/README.md similarity index 100% rename from parakeet/frontend/zh_normalization/README.md rename to paddlespeech/t2s/frontend/zh_normalization/README.md diff --git a/paddlespeech/t2s/frontend/zh_normalization/__init__.py b/paddlespeech/t2s/frontend/zh_normalization/__init__.py new file mode 100644 index 00000000..a9d1f44d --- /dev/null +++ b/paddlespeech/t2s/frontend/zh_normalization/__init__.py @@ -0,0 +1,14 @@ +# Copyright (c) 2020 PaddlePaddle Authors. All Rights Reserved. +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +from paddlespeech.t2s.frontend.zh_normalization.text_normlization import * diff --git a/parakeet/frontend/zh_normalization/char_convert.py b/paddlespeech/t2s/frontend/zh_normalization/char_convert.py similarity index 100% rename from parakeet/frontend/zh_normalization/char_convert.py rename to paddlespeech/t2s/frontend/zh_normalization/char_convert.py diff --git a/parakeet/frontend/zh_normalization/chronology.py b/paddlespeech/t2s/frontend/zh_normalization/chronology.py similarity index 100% rename from parakeet/frontend/zh_normalization/chronology.py rename to paddlespeech/t2s/frontend/zh_normalization/chronology.py diff --git a/parakeet/frontend/zh_normalization/constants.py b/paddlespeech/t2s/frontend/zh_normalization/constants.py similarity index 100% rename from parakeet/frontend/zh_normalization/constants.py rename to paddlespeech/t2s/frontend/zh_normalization/constants.py diff --git a/parakeet/frontend/zh_normalization/num.py b/paddlespeech/t2s/frontend/zh_normalization/num.py similarity index 100% rename from parakeet/frontend/zh_normalization/num.py rename to paddlespeech/t2s/frontend/zh_normalization/num.py diff --git a/parakeet/frontend/zh_normalization/phonecode.py b/paddlespeech/t2s/frontend/zh_normalization/phonecode.py similarity index 100% rename from parakeet/frontend/zh_normalization/phonecode.py rename to paddlespeech/t2s/frontend/zh_normalization/phonecode.py diff --git a/parakeet/frontend/zh_normalization/quantifier.py b/paddlespeech/t2s/frontend/zh_normalization/quantifier.py similarity index 100% rename from parakeet/frontend/zh_normalization/quantifier.py rename to paddlespeech/t2s/frontend/zh_normalization/quantifier.py diff --git a/parakeet/frontend/zh_normalization/text_normlization.py b/paddlespeech/t2s/frontend/zh_normalization/text_normlization.py similarity index 100% rename from parakeet/frontend/zh_normalization/text_normlization.py rename to paddlespeech/t2s/frontend/zh_normalization/text_normlization.py diff --git a/parakeet/models/__init__.py b/paddlespeech/t2s/models/__init__.py similarity index 100% rename from parakeet/models/__init__.py rename to paddlespeech/t2s/models/__init__.py diff --git a/parakeet/models/fastspeech2/__init__.py b/paddlespeech/t2s/models/fastspeech2/__init__.py similarity index 100% rename from parakeet/models/fastspeech2/__init__.py rename to paddlespeech/t2s/models/fastspeech2/__init__.py diff --git a/parakeet/models/fastspeech2/fastspeech2.py b/paddlespeech/t2s/models/fastspeech2/fastspeech2.py similarity index 97% rename from parakeet/models/fastspeech2/fastspeech2.py rename to paddlespeech/t2s/models/fastspeech2/fastspeech2.py index 192517b1..754fef2a 100644 --- a/parakeet/models/fastspeech2/fastspeech2.py +++ b/paddlespeech/t2s/models/fastspeech2/fastspeech2.py @@ -21,17 +21,17 @@ import paddle.nn.functional as F from paddle import nn from typeguard import check_argument_types -from parakeet.modules.fastspeech2_predictor.duration_predictor import DurationPredictor -from parakeet.modules.fastspeech2_predictor.duration_predictor import DurationPredictorLoss -from parakeet.modules.fastspeech2_predictor.length_regulator import LengthRegulator -from parakeet.modules.fastspeech2_predictor.variance_predictor import VariancePredictor -from parakeet.modules.fastspeech2_transformer.embedding import PositionalEncoding -from parakeet.modules.fastspeech2_transformer.embedding import ScaledPositionalEncoding -from parakeet.modules.fastspeech2_transformer.encoder import Encoder as TransformerEncoder -from parakeet.modules.nets_utils import initialize -from parakeet.modules.nets_utils import make_non_pad_mask -from parakeet.modules.nets_utils import make_pad_mask -from parakeet.modules.tacotron2.decoder import Postnet +from paddlespeech.t2s.modules.fastspeech2_predictor.duration_predictor import DurationPredictor +from paddlespeech.t2s.modules.fastspeech2_predictor.duration_predictor import DurationPredictorLoss +from paddlespeech.t2s.modules.fastspeech2_predictor.length_regulator import LengthRegulator +from paddlespeech.t2s.modules.fastspeech2_predictor.variance_predictor import VariancePredictor +from paddlespeech.t2s.modules.fastspeech2_transformer.embedding import PositionalEncoding +from paddlespeech.t2s.modules.fastspeech2_transformer.embedding import ScaledPositionalEncoding +from paddlespeech.t2s.modules.fastspeech2_transformer.encoder import Encoder as TransformerEncoder +from paddlespeech.t2s.modules.nets_utils import initialize +from paddlespeech.t2s.modules.nets_utils import make_non_pad_mask +from paddlespeech.t2s.modules.nets_utils import make_pad_mask +from paddlespeech.t2s.modules.tacotron2.decoder import Postnet class FastSpeech2(nn.Layer): diff --git a/parakeet/models/fastspeech2/fastspeech2_updater.py b/paddlespeech/t2s/models/fastspeech2/fastspeech2_updater.py similarity index 95% rename from parakeet/models/fastspeech2/fastspeech2_updater.py rename to paddlespeech/t2s/models/fastspeech2/fastspeech2_updater.py index ea23ec2a..4297c8b6 100644 --- a/parakeet/models/fastspeech2/fastspeech2_updater.py +++ b/paddlespeech/t2s/models/fastspeech2/fastspeech2_updater.py @@ -15,10 +15,10 @@ import logging from paddle import distributed as dist -from parakeet.models.fastspeech2 import FastSpeech2Loss -from parakeet.training.extensions.evaluator import StandardEvaluator -from parakeet.training.reporter import report -from parakeet.training.updaters.standard_updater import StandardUpdater +from paddlespeech.t2s.models.fastspeech2 import FastSpeech2Loss +from paddlespeech.t2s.training.extensions.evaluator import StandardEvaluator +from paddlespeech.t2s.training.reporter import report +from paddlespeech.t2s.training.updaters.standard_updater import StandardUpdater logging.basicConfig( format='%(asctime)s [%(levelname)s] [%(filename)s:%(lineno)d] %(message)s', datefmt='[%Y-%m-%d %H:%M:%S]') diff --git a/parakeet/models/lstm_speaker_encoder.py b/paddlespeech/t2s/models/lstm_speaker_encoder.py similarity index 100% rename from parakeet/models/lstm_speaker_encoder.py rename to paddlespeech/t2s/models/lstm_speaker_encoder.py diff --git a/parakeet/models/melgan/__init__.py b/paddlespeech/t2s/models/melgan/__init__.py similarity index 100% rename from parakeet/models/melgan/__init__.py rename to paddlespeech/t2s/models/melgan/__init__.py diff --git a/parakeet/models/melgan/melgan.py b/paddlespeech/t2s/models/melgan/melgan.py similarity index 98% rename from parakeet/models/melgan/melgan.py rename to paddlespeech/t2s/models/melgan/melgan.py index 5de8a298..94d6e631 100644 --- a/parakeet/models/melgan/melgan.py +++ b/paddlespeech/t2s/models/melgan/melgan.py @@ -20,11 +20,11 @@ import numpy as np import paddle from paddle import nn -from parakeet.modules.causal_conv import CausalConv1D -from parakeet.modules.causal_conv import CausalConv1DTranspose -from parakeet.modules.nets_utils import initialize -from parakeet.modules.pqmf import PQMF -from parakeet.modules.residual_stack import ResidualStack +from paddlespeech.t2s.modules.causal_conv import CausalConv1D +from paddlespeech.t2s.modules.causal_conv import CausalConv1DTranspose +from paddlespeech.t2s.modules.nets_utils import initialize +from paddlespeech.t2s.modules.pqmf import PQMF +from paddlespeech.t2s.modules.residual_stack import ResidualStack class MelGANGenerator(nn.Layer): diff --git a/parakeet/models/melgan/multi_band_melgan_updater.py b/paddlespeech/t2s/models/melgan/multi_band_melgan_updater.py similarity index 97% rename from parakeet/models/melgan/multi_band_melgan_updater.py rename to paddlespeech/t2s/models/melgan/multi_band_melgan_updater.py index 0783cb97..a5d4cdeb 100644 --- a/parakeet/models/melgan/multi_band_melgan_updater.py +++ b/paddlespeech/t2s/models/melgan/multi_band_melgan_updater.py @@ -21,10 +21,10 @@ from paddle.nn import Layer from paddle.optimizer import Optimizer from paddle.optimizer.lr import LRScheduler -from parakeet.training.extensions.evaluator import StandardEvaluator -from parakeet.training.reporter import report -from parakeet.training.updaters.standard_updater import StandardUpdater -from parakeet.training.updaters.standard_updater import UpdaterState +from paddlespeech.t2s.training.extensions.evaluator import StandardEvaluator +from paddlespeech.t2s.training.reporter import report +from paddlespeech.t2s.training.updaters.standard_updater import StandardUpdater +from paddlespeech.t2s.training.updaters.standard_updater import UpdaterState logging.basicConfig( format='%(asctime)s [%(levelname)s] [%(filename)s:%(lineno)d] %(message)s', datefmt='[%Y-%m-%d %H:%M:%S]') diff --git a/parakeet/models/parallel_wavegan/__init__.py b/paddlespeech/t2s/models/parallel_wavegan/__init__.py similarity index 100% rename from parakeet/models/parallel_wavegan/__init__.py rename to paddlespeech/t2s/models/parallel_wavegan/__init__.py diff --git a/parakeet/models/parallel_wavegan/parallel_wavegan.py b/paddlespeech/t2s/models/parallel_wavegan/parallel_wavegan.py similarity index 100% rename from parakeet/models/parallel_wavegan/parallel_wavegan.py rename to paddlespeech/t2s/models/parallel_wavegan/parallel_wavegan.py diff --git a/parakeet/models/parallel_wavegan/parallel_wavegan_updater.py b/paddlespeech/t2s/models/parallel_wavegan/parallel_wavegan_updater.py similarity index 96% rename from parakeet/models/parallel_wavegan/parallel_wavegan_updater.py rename to paddlespeech/t2s/models/parallel_wavegan/parallel_wavegan_updater.py index 7bd59881..4e3daaa3 100644 --- a/parakeet/models/parallel_wavegan/parallel_wavegan_updater.py +++ b/paddlespeech/t2s/models/parallel_wavegan/parallel_wavegan_updater.py @@ -22,10 +22,10 @@ from paddle.optimizer import Optimizer from paddle.optimizer.lr import LRScheduler from timer import timer -from parakeet.training.extensions.evaluator import StandardEvaluator -from parakeet.training.reporter import report -from parakeet.training.updaters.standard_updater import StandardUpdater -from parakeet.training.updaters.standard_updater import UpdaterState +from paddlespeech.t2s.training.extensions.evaluator import StandardEvaluator +from paddlespeech.t2s.training.reporter import report +from paddlespeech.t2s.training.updaters.standard_updater import StandardUpdater +from paddlespeech.t2s.training.updaters.standard_updater import UpdaterState logging.basicConfig( format='%(asctime)s [%(levelname)s] [%(filename)s:%(lineno)d] %(message)s', datefmt='[%Y-%m-%d %H:%M:%S]') diff --git a/parakeet/models/speedyspeech/__init__.py b/paddlespeech/t2s/models/speedyspeech/__init__.py similarity index 100% rename from parakeet/models/speedyspeech/__init__.py rename to paddlespeech/t2s/models/speedyspeech/__init__.py diff --git a/parakeet/models/speedyspeech/speedyspeech.py b/paddlespeech/t2s/models/speedyspeech/speedyspeech.py similarity index 98% rename from parakeet/models/speedyspeech/speedyspeech.py rename to paddlespeech/t2s/models/speedyspeech/speedyspeech.py index dea567b6..0689ec45 100644 --- a/parakeet/models/speedyspeech/speedyspeech.py +++ b/paddlespeech/t2s/models/speedyspeech/speedyspeech.py @@ -14,8 +14,8 @@ import paddle from paddle import nn -from parakeet.modules.expansion import expand -from parakeet.modules.positional_encoding import sinusoid_position_encoding +from paddlespeech.t2s.modules.expansion import expand +from paddlespeech.t2s.modules.positional_encoding import sinusoid_position_encoding class ResidualBlock(nn.Layer): diff --git a/parakeet/models/speedyspeech/speedyspeech_updater.py b/paddlespeech/t2s/models/speedyspeech/speedyspeech_updater.py similarity index 93% rename from parakeet/models/speedyspeech/speedyspeech_updater.py rename to paddlespeech/t2s/models/speedyspeech/speedyspeech_updater.py index a17c93c7..4883a87e 100644 --- a/parakeet/models/speedyspeech/speedyspeech_updater.py +++ b/paddlespeech/t2s/models/speedyspeech/speedyspeech_updater.py @@ -18,12 +18,12 @@ from paddle import distributed as dist from paddle.fluid.layers import huber_loss from paddle.nn import functional as F -from parakeet.modules.losses import masked_l1_loss -from parakeet.modules.losses import weighted_mean -from parakeet.modules.ssim import ssim -from parakeet.training.extensions.evaluator import StandardEvaluator -from parakeet.training.reporter import report -from parakeet.training.updaters.standard_updater import StandardUpdater +from paddlespeech.t2s.modules.losses import masked_l1_loss +from paddlespeech.t2s.modules.losses import weighted_mean +from paddlespeech.t2s.modules.ssim import ssim +from paddlespeech.t2s.training.extensions.evaluator import StandardEvaluator +from paddlespeech.t2s.training.reporter import report +from paddlespeech.t2s.training.updaters.standard_updater import StandardUpdater logging.basicConfig( format='%(asctime)s [%(levelname)s] [%(filename)s:%(lineno)d] %(message)s', datefmt='[%Y-%m-%d %H:%M:%S]') diff --git a/parakeet/models/tacotron2.py b/paddlespeech/t2s/models/tacotron2.py similarity index 99% rename from parakeet/models/tacotron2.py rename to paddlespeech/t2s/models/tacotron2.py index 07af6b3f..b0946a5b 100644 --- a/parakeet/models/tacotron2.py +++ b/paddlespeech/t2s/models/tacotron2.py @@ -20,10 +20,10 @@ from paddle.nn import functional as F from paddle.nn import initializer as I from tqdm import trange -from parakeet.modules.attention import LocationSensitiveAttention -from parakeet.modules.conv import Conv1dBatchNorm -from parakeet.modules.losses import guided_attention_loss -from parakeet.utils import checkpoint +from paddlespeech.t2s.modules.attention import LocationSensitiveAttention +from paddlespeech.t2s.modules.conv import Conv1dBatchNorm +from paddlespeech.t2s.modules.losses import guided_attention_loss +from paddlespeech.t2s.utils import checkpoint __all__ = ["Tacotron2", "Tacotron2Loss"] diff --git a/parakeet/models/transformer_tts/__init__.py b/paddlespeech/t2s/models/transformer_tts/__init__.py similarity index 100% rename from parakeet/models/transformer_tts/__init__.py rename to paddlespeech/t2s/models/transformer_tts/__init__.py diff --git a/parakeet/models/transformer_tts/transformer_tts.py b/paddlespeech/t2s/models/transformer_tts/transformer_tts.py similarity index 97% rename from parakeet/models/transformer_tts/transformer_tts.py rename to paddlespeech/t2s/models/transformer_tts/transformer_tts.py index bb3674f3..f8feb553 100644 --- a/parakeet/models/transformer_tts/transformer_tts.py +++ b/paddlespeech/t2s/models/transformer_tts/transformer_tts.py @@ -22,19 +22,19 @@ import paddle.nn.functional as F from paddle import nn from typeguard import check_argument_types -from parakeet.modules.fastspeech2_transformer.attention import MultiHeadedAttention -from parakeet.modules.fastspeech2_transformer.decoder import Decoder -from parakeet.modules.fastspeech2_transformer.embedding import PositionalEncoding -from parakeet.modules.fastspeech2_transformer.embedding import ScaledPositionalEncoding -from parakeet.modules.fastspeech2_transformer.encoder import Encoder -from parakeet.modules.fastspeech2_transformer.mask import subsequent_mask -from parakeet.modules.nets_utils import initialize -from parakeet.modules.nets_utils import make_non_pad_mask -from parakeet.modules.nets_utils import make_pad_mask -from parakeet.modules.style_encoder import StyleEncoder -from parakeet.modules.tacotron2.decoder import Postnet -from parakeet.modules.tacotron2.decoder import Prenet as DecoderPrenet -from parakeet.modules.tacotron2.encoder import Encoder as EncoderPrenet +from paddlespeech.t2s.modules.fastspeech2_transformer.attention import MultiHeadedAttention +from paddlespeech.t2s.modules.fastspeech2_transformer.decoder import Decoder +from paddlespeech.t2s.modules.fastspeech2_transformer.embedding import PositionalEncoding +from paddlespeech.t2s.modules.fastspeech2_transformer.embedding import ScaledPositionalEncoding +from paddlespeech.t2s.modules.fastspeech2_transformer.encoder import Encoder +from paddlespeech.t2s.modules.fastspeech2_transformer.mask import subsequent_mask +from paddlespeech.t2s.modules.nets_utils import initialize +from paddlespeech.t2s.modules.nets_utils import make_non_pad_mask +from paddlespeech.t2s.modules.nets_utils import make_pad_mask +from paddlespeech.t2s.modules.style_encoder import StyleEncoder +from paddlespeech.t2s.modules.tacotron2.decoder import Postnet +from paddlespeech.t2s.modules.tacotron2.decoder import Prenet as DecoderPrenet +from paddlespeech.t2s.modules.tacotron2.encoder import Encoder as EncoderPrenet class TransformerTTS(nn.Layer): diff --git a/parakeet/models/transformer_tts/transformer_tts_updater.py b/paddlespeech/t2s/models/transformer_tts/transformer_tts_updater.py similarity index 97% rename from parakeet/models/transformer_tts/transformer_tts_updater.py rename to paddlespeech/t2s/models/transformer_tts/transformer_tts_updater.py index 4bec4758..f16cf4dd 100644 --- a/parakeet/models/transformer_tts/transformer_tts_updater.py +++ b/paddlespeech/t2s/models/transformer_tts/transformer_tts_updater.py @@ -17,11 +17,11 @@ from typing import Sequence import paddle from paddle import distributed as dist -from parakeet.models.transformer_tts import GuidedMultiHeadAttentionLoss -from parakeet.models.transformer_tts import TransformerTTSLoss -from parakeet.training.extensions.evaluator import StandardEvaluator -from parakeet.training.reporter import report -from parakeet.training.updaters.standard_updater import StandardUpdater +from paddlespeech.t2s.models.transformer_tts import GuidedMultiHeadAttentionLoss +from paddlespeech.t2s.models.transformer_tts import TransformerTTSLoss +from paddlespeech.t2s.training.extensions.evaluator import StandardEvaluator +from paddlespeech.t2s.training.reporter import report +from paddlespeech.t2s.training.updaters.standard_updater import StandardUpdater logging.basicConfig( format='%(asctime)s [%(levelname)s] [%(filename)s:%(lineno)d] %(message)s', datefmt='[%Y-%m-%d %H:%M:%S]') diff --git a/parakeet/models/waveflow.py b/paddlespeech/t2s/models/waveflow.py similarity index 99% rename from parakeet/models/waveflow.py rename to paddlespeech/t2s/models/waveflow.py index b6317bf8..c57429db 100644 --- a/parakeet/models/waveflow.py +++ b/paddlespeech/t2s/models/waveflow.py @@ -23,8 +23,8 @@ from paddle import nn from paddle.nn import functional as F from paddle.nn import initializer as I -from parakeet.modules import geometry as geo -from parakeet.utils import checkpoint +from paddlespeech.t2s.modules import geometry as geo +from paddlespeech.t2s.utils import checkpoint __all__ = ["WaveFlow", "ConditionalWaveFlow", "WaveFlowLoss"] diff --git a/parakeet/modules/__init__.py b/paddlespeech/t2s/modules/__init__.py similarity index 100% rename from parakeet/modules/__init__.py rename to paddlespeech/t2s/modules/__init__.py diff --git a/parakeet/modules/adversarial_loss.py b/paddlespeech/t2s/modules/adversarial_loss.py similarity index 100% rename from parakeet/modules/adversarial_loss.py rename to paddlespeech/t2s/modules/adversarial_loss.py diff --git a/parakeet/modules/attention.py b/paddlespeech/t2s/modules/attention.py similarity index 100% rename from parakeet/modules/attention.py rename to paddlespeech/t2s/modules/attention.py diff --git a/parakeet/modules/audio.py b/paddlespeech/t2s/modules/audio.py similarity index 100% rename from parakeet/modules/audio.py rename to paddlespeech/t2s/modules/audio.py diff --git a/parakeet/modules/causal_conv.py b/paddlespeech/t2s/modules/causal_conv.py similarity index 100% rename from parakeet/modules/causal_conv.py rename to paddlespeech/t2s/modules/causal_conv.py diff --git a/parakeet/modules/conv.py b/paddlespeech/t2s/modules/conv.py similarity index 100% rename from parakeet/modules/conv.py rename to paddlespeech/t2s/modules/conv.py diff --git a/parakeet/modules/expansion.py b/paddlespeech/t2s/modules/expansion.py similarity index 100% rename from parakeet/modules/expansion.py rename to paddlespeech/t2s/modules/expansion.py diff --git a/parakeet/modules/fastspeech2_predictor/__init__.py b/paddlespeech/t2s/modules/fastspeech2_predictor/__init__.py similarity index 100% rename from parakeet/modules/fastspeech2_predictor/__init__.py rename to paddlespeech/t2s/modules/fastspeech2_predictor/__init__.py diff --git a/parakeet/modules/fastspeech2_predictor/duration_predictor.py b/paddlespeech/t2s/modules/fastspeech2_predictor/duration_predictor.py similarity index 97% rename from parakeet/modules/fastspeech2_predictor/duration_predictor.py rename to paddlespeech/t2s/modules/fastspeech2_predictor/duration_predictor.py index d0f5262c..eac7cf99 100644 --- a/parakeet/modules/fastspeech2_predictor/duration_predictor.py +++ b/paddlespeech/t2s/modules/fastspeech2_predictor/duration_predictor.py @@ -15,8 +15,8 @@ import paddle from paddle import nn -from parakeet.modules.layer_norm import LayerNorm -from parakeet.modules.masked_fill import masked_fill +from paddlespeech.t2s.modules.layer_norm import LayerNorm +from paddlespeech.t2s.modules.masked_fill import masked_fill class DurationPredictor(nn.Layer): diff --git a/parakeet/modules/fastspeech2_predictor/length_regulator.py b/paddlespeech/t2s/modules/fastspeech2_predictor/length_regulator.py similarity index 100% rename from parakeet/modules/fastspeech2_predictor/length_regulator.py rename to paddlespeech/t2s/modules/fastspeech2_predictor/length_regulator.py diff --git a/parakeet/modules/fastspeech2_predictor/variance_predictor.py b/paddlespeech/t2s/modules/fastspeech2_predictor/variance_predictor.py similarity index 96% rename from parakeet/modules/fastspeech2_predictor/variance_predictor.py rename to paddlespeech/t2s/modules/fastspeech2_predictor/variance_predictor.py index 0a980dd5..0fd108d3 100644 --- a/parakeet/modules/fastspeech2_predictor/variance_predictor.py +++ b/paddlespeech/t2s/modules/fastspeech2_predictor/variance_predictor.py @@ -16,8 +16,8 @@ import paddle from paddle import nn from typeguard import check_argument_types -from parakeet.modules.layer_norm import LayerNorm -from parakeet.modules.masked_fill import masked_fill +from paddlespeech.t2s.modules.layer_norm import LayerNorm +from paddlespeech.t2s.modules.masked_fill import masked_fill class VariancePredictor(nn.Layer): diff --git a/parakeet/modules/fastspeech2_transformer/__init__.py b/paddlespeech/t2s/modules/fastspeech2_transformer/__init__.py similarity index 100% rename from parakeet/modules/fastspeech2_transformer/__init__.py rename to paddlespeech/t2s/modules/fastspeech2_transformer/__init__.py diff --git a/parakeet/modules/fastspeech2_transformer/attention.py b/paddlespeech/t2s/modules/fastspeech2_transformer/attention.py similarity index 98% rename from parakeet/modules/fastspeech2_transformer/attention.py rename to paddlespeech/t2s/modules/fastspeech2_transformer/attention.py index 0bac4742..a61f1245 100644 --- a/parakeet/modules/fastspeech2_transformer/attention.py +++ b/paddlespeech/t2s/modules/fastspeech2_transformer/attention.py @@ -18,7 +18,7 @@ import numpy import paddle from paddle import nn -from parakeet.modules.masked_fill import masked_fill +from paddlespeech.t2s.modules.masked_fill import masked_fill class MultiHeadedAttention(nn.Layer): diff --git a/parakeet/modules/fastspeech2_transformer/decoder.py b/paddlespeech/t2s/modules/fastspeech2_transformer/decoder.py similarity index 93% rename from parakeet/modules/fastspeech2_transformer/decoder.py rename to paddlespeech/t2s/modules/fastspeech2_transformer/decoder.py index 0f09014f..c41e3995 100644 --- a/parakeet/modules/fastspeech2_transformer/decoder.py +++ b/paddlespeech/t2s/modules/fastspeech2_transformer/decoder.py @@ -22,14 +22,14 @@ import paddle import paddle.nn.functional as F from paddle import nn -from parakeet.modules.fastspeech2_transformer.attention import MultiHeadedAttention -from parakeet.modules.fastspeech2_transformer.decoder_layer import DecoderLayer -from parakeet.modules.fastspeech2_transformer.embedding import PositionalEncoding -from parakeet.modules.fastspeech2_transformer.lightconv import LightweightConvolution -from parakeet.modules.fastspeech2_transformer.mask import subsequent_mask -from parakeet.modules.fastspeech2_transformer.positionwise_feed_forward import PositionwiseFeedForward -from parakeet.modules.fastspeech2_transformer.repeat import repeat -from parakeet.modules.layer_norm import LayerNorm +from paddlespeech.t2s.modules.fastspeech2_transformer.attention import MultiHeadedAttention +from paddlespeech.t2s.modules.fastspeech2_transformer.decoder_layer import DecoderLayer +from paddlespeech.t2s.modules.fastspeech2_transformer.embedding import PositionalEncoding +from paddlespeech.t2s.modules.fastspeech2_transformer.lightconv import LightweightConvolution +from paddlespeech.t2s.modules.fastspeech2_transformer.mask import subsequent_mask +from paddlespeech.t2s.modules.fastspeech2_transformer.positionwise_feed_forward import PositionwiseFeedForward +from paddlespeech.t2s.modules.fastspeech2_transformer.repeat import repeat +from paddlespeech.t2s.modules.layer_norm import LayerNorm class Decoder(nn.Layer): diff --git a/parakeet/modules/fastspeech2_transformer/decoder_layer.py b/paddlespeech/t2s/modules/fastspeech2_transformer/decoder_layer.py similarity index 98% rename from parakeet/modules/fastspeech2_transformer/decoder_layer.py rename to paddlespeech/t2s/modules/fastspeech2_transformer/decoder_layer.py index f968051e..9b042483 100644 --- a/parakeet/modules/fastspeech2_transformer/decoder_layer.py +++ b/paddlespeech/t2s/modules/fastspeech2_transformer/decoder_layer.py @@ -15,7 +15,7 @@ import paddle from paddle import nn -from parakeet.modules.layer_norm import LayerNorm +from paddlespeech.t2s.modules.layer_norm import LayerNorm class DecoderLayer(nn.Layer): diff --git a/parakeet/modules/fastspeech2_transformer/embedding.py b/paddlespeech/t2s/modules/fastspeech2_transformer/embedding.py similarity index 100% rename from parakeet/modules/fastspeech2_transformer/embedding.py rename to paddlespeech/t2s/modules/fastspeech2_transformer/embedding.py diff --git a/parakeet/modules/fastspeech2_transformer/encoder.py b/paddlespeech/t2s/modules/fastspeech2_transformer/encoder.py similarity index 92% rename from parakeet/modules/fastspeech2_transformer/encoder.py rename to paddlespeech/t2s/modules/fastspeech2_transformer/encoder.py index 996e9dee..904b9d8b 100644 --- a/parakeet/modules/fastspeech2_transformer/encoder.py +++ b/paddlespeech/t2s/modules/fastspeech2_transformer/encoder.py @@ -15,13 +15,13 @@ import logging from paddle import nn -from parakeet.modules.fastspeech2_transformer.attention import MultiHeadedAttention -from parakeet.modules.fastspeech2_transformer.embedding import PositionalEncoding -from parakeet.modules.fastspeech2_transformer.encoder_layer import EncoderLayer -from parakeet.modules.fastspeech2_transformer.multi_layer_conv import Conv1dLinear -from parakeet.modules.fastspeech2_transformer.multi_layer_conv import MultiLayeredConv1d -from parakeet.modules.fastspeech2_transformer.positionwise_feed_forward import PositionwiseFeedForward -from parakeet.modules.fastspeech2_transformer.repeat import repeat +from paddlespeech.t2s.modules.fastspeech2_transformer.attention import MultiHeadedAttention +from paddlespeech.t2s.modules.fastspeech2_transformer.embedding import PositionalEncoding +from paddlespeech.t2s.modules.fastspeech2_transformer.encoder_layer import EncoderLayer +from paddlespeech.t2s.modules.fastspeech2_transformer.multi_layer_conv import Conv1dLinear +from paddlespeech.t2s.modules.fastspeech2_transformer.multi_layer_conv import MultiLayeredConv1d +from paddlespeech.t2s.modules.fastspeech2_transformer.positionwise_feed_forward import PositionwiseFeedForward +from paddlespeech.t2s.modules.fastspeech2_transformer.repeat import repeat class Encoder(nn.Layer): diff --git a/parakeet/modules/fastspeech2_transformer/encoder_layer.py b/paddlespeech/t2s/modules/fastspeech2_transformer/encoder_layer.py similarity index 100% rename from parakeet/modules/fastspeech2_transformer/encoder_layer.py rename to paddlespeech/t2s/modules/fastspeech2_transformer/encoder_layer.py diff --git a/parakeet/modules/fastspeech2_transformer/lightconv.py b/paddlespeech/t2s/modules/fastspeech2_transformer/lightconv.py similarity index 97% rename from parakeet/modules/fastspeech2_transformer/lightconv.py rename to paddlespeech/t2s/modules/fastspeech2_transformer/lightconv.py index 06116884..3aceed31 100644 --- a/parakeet/modules/fastspeech2_transformer/lightconv.py +++ b/paddlespeech/t2s/modules/fastspeech2_transformer/lightconv.py @@ -17,8 +17,8 @@ import paddle import paddle.nn.functional as F from paddle import nn -from parakeet.modules.glu import GLU -from parakeet.modules.masked_fill import masked_fill +from paddlespeech.t2s.modules.glu import GLU +from paddlespeech.t2s.modules.masked_fill import masked_fill MIN_VALUE = float(numpy.finfo(numpy.float32).min) diff --git a/parakeet/modules/fastspeech2_transformer/mask.py b/paddlespeech/t2s/modules/fastspeech2_transformer/mask.py similarity index 100% rename from parakeet/modules/fastspeech2_transformer/mask.py rename to paddlespeech/t2s/modules/fastspeech2_transformer/mask.py diff --git a/parakeet/modules/fastspeech2_transformer/multi_layer_conv.py b/paddlespeech/t2s/modules/fastspeech2_transformer/multi_layer_conv.py similarity index 100% rename from parakeet/modules/fastspeech2_transformer/multi_layer_conv.py rename to paddlespeech/t2s/modules/fastspeech2_transformer/multi_layer_conv.py diff --git a/parakeet/modules/fastspeech2_transformer/positionwise_feed_forward.py b/paddlespeech/t2s/modules/fastspeech2_transformer/positionwise_feed_forward.py similarity index 100% rename from parakeet/modules/fastspeech2_transformer/positionwise_feed_forward.py rename to paddlespeech/t2s/modules/fastspeech2_transformer/positionwise_feed_forward.py diff --git a/parakeet/modules/fastspeech2_transformer/repeat.py b/paddlespeech/t2s/modules/fastspeech2_transformer/repeat.py similarity index 100% rename from parakeet/modules/fastspeech2_transformer/repeat.py rename to paddlespeech/t2s/modules/fastspeech2_transformer/repeat.py diff --git a/parakeet/modules/geometry.py b/paddlespeech/t2s/modules/geometry.py similarity index 100% rename from parakeet/modules/geometry.py rename to paddlespeech/t2s/modules/geometry.py diff --git a/parakeet/modules/glu.py b/paddlespeech/t2s/modules/glu.py similarity index 100% rename from parakeet/modules/glu.py rename to paddlespeech/t2s/modules/glu.py diff --git a/parakeet/modules/layer_norm.py b/paddlespeech/t2s/modules/layer_norm.py similarity index 100% rename from parakeet/modules/layer_norm.py rename to paddlespeech/t2s/modules/layer_norm.py diff --git a/parakeet/modules/losses.py b/paddlespeech/t2s/modules/losses.py similarity index 100% rename from parakeet/modules/losses.py rename to paddlespeech/t2s/modules/losses.py diff --git a/parakeet/modules/masked_fill.py b/paddlespeech/t2s/modules/masked_fill.py similarity index 100% rename from parakeet/modules/masked_fill.py rename to paddlespeech/t2s/modules/masked_fill.py diff --git a/parakeet/modules/masking.py b/paddlespeech/t2s/modules/masking.py similarity index 100% rename from parakeet/modules/masking.py rename to paddlespeech/t2s/modules/masking.py diff --git a/parakeet/modules/nets_utils.py b/paddlespeech/t2s/modules/nets_utils.py similarity index 100% rename from parakeet/modules/nets_utils.py rename to paddlespeech/t2s/modules/nets_utils.py diff --git a/parakeet/modules/normalizer.py b/paddlespeech/t2s/modules/normalizer.py similarity index 100% rename from parakeet/modules/normalizer.py rename to paddlespeech/t2s/modules/normalizer.py diff --git a/parakeet/modules/positional_encoding.py b/paddlespeech/t2s/modules/positional_encoding.py similarity index 100% rename from parakeet/modules/positional_encoding.py rename to paddlespeech/t2s/modules/positional_encoding.py diff --git a/parakeet/modules/pqmf.py b/paddlespeech/t2s/modules/pqmf.py similarity index 100% rename from parakeet/modules/pqmf.py rename to paddlespeech/t2s/modules/pqmf.py diff --git a/parakeet/modules/residual_stack.py b/paddlespeech/t2s/modules/residual_stack.py similarity index 98% rename from parakeet/modules/residual_stack.py rename to paddlespeech/t2s/modules/residual_stack.py index 135c32e5..71bd4f0c 100644 --- a/parakeet/modules/residual_stack.py +++ b/paddlespeech/t2s/modules/residual_stack.py @@ -17,7 +17,7 @@ from typing import Dict from paddle import nn -from parakeet.modules.causal_conv import CausalConv1D +from paddlespeech.t2s.modules.causal_conv import CausalConv1D class ResidualStack(nn.Layer): diff --git a/parakeet/modules/ssim.py b/paddlespeech/t2s/modules/ssim.py similarity index 100% rename from parakeet/modules/ssim.py rename to paddlespeech/t2s/modules/ssim.py diff --git a/parakeet/modules/stft_loss.py b/paddlespeech/t2s/modules/stft_loss.py similarity index 100% rename from parakeet/modules/stft_loss.py rename to paddlespeech/t2s/modules/stft_loss.py diff --git a/parakeet/modules/style_encoder.py b/paddlespeech/t2s/modules/style_encoder.py similarity index 98% rename from parakeet/modules/style_encoder.py rename to paddlespeech/t2s/modules/style_encoder.py index fb27258c..c6b12d44 100644 --- a/parakeet/modules/style_encoder.py +++ b/paddlespeech/t2s/modules/style_encoder.py @@ -18,7 +18,7 @@ import paddle from paddle import nn from typeguard import check_argument_types -from parakeet.modules.fastspeech2_transformer.attention import MultiHeadedAttention as BaseMultiHeadedAttention +from paddlespeech.t2s.modules.fastspeech2_transformer.attention import MultiHeadedAttention as BaseMultiHeadedAttention class StyleEncoder(nn.Layer): diff --git a/parakeet/modules/tacotron2/__init__.py b/paddlespeech/t2s/modules/tacotron2/__init__.py similarity index 100% rename from parakeet/modules/tacotron2/__init__.py rename to paddlespeech/t2s/modules/tacotron2/__init__.py diff --git a/parakeet/modules/tacotron2/decoder.py b/paddlespeech/t2s/modules/tacotron2/decoder.py similarity index 100% rename from parakeet/modules/tacotron2/decoder.py rename to paddlespeech/t2s/modules/tacotron2/decoder.py diff --git a/parakeet/modules/tacotron2/encoder.py b/paddlespeech/t2s/modules/tacotron2/encoder.py similarity index 100% rename from parakeet/modules/tacotron2/encoder.py rename to paddlespeech/t2s/modules/tacotron2/encoder.py diff --git a/parakeet/modules/transformer.py b/paddlespeech/t2s/modules/transformer.py similarity index 99% rename from parakeet/modules/transformer.py rename to paddlespeech/t2s/modules/transformer.py index 490458be..e50d58d4 100644 --- a/parakeet/modules/transformer.py +++ b/paddlespeech/t2s/modules/transformer.py @@ -14,7 +14,7 @@ from paddle import nn from paddle.nn import functional as F -from parakeet.modules import attention as attn +from paddlespeech.t2s.modules import attention as attn __all__ = [ "PositionwiseFFN", diff --git a/parakeet/training/__init__.py b/paddlespeech/t2s/training/__init__.py similarity index 100% rename from parakeet/training/__init__.py rename to paddlespeech/t2s/training/__init__.py diff --git a/parakeet/training/cli.py b/paddlespeech/t2s/training/cli.py similarity index 91% rename from parakeet/training/cli.py rename to paddlespeech/t2s/training/cli.py index a6309941..3b9fd42e 100644 --- a/parakeet/training/cli.py +++ b/paddlespeech/t2s/training/cli.py @@ -15,10 +15,10 @@ import argparse def default_argument_parser(): - r"""A simple yet genral argument parser for experiments with parakeet. + r"""A simple yet genral argument parser for experiments with t2s. - This is used in examples with parakeet. And it is intended to be used by - other experiments with parakeet. It requires a minimal set of command line + This is used in examples with t2s. And it is intended to be used by + other experiments with t2s. It requires a minimal set of command line arguments to start a training script. The ``--config`` and ``--opts`` are used for overwrite the deault @@ -34,7 +34,7 @@ def default_argument_parser(): See Also -------- - parakeet.training.experiment + paddlespeech.t2s.training.experiment Returns ------- diff --git a/parakeet/training/default_config.py b/paddlespeech/t2s/training/default_config.py similarity index 100% rename from parakeet/training/default_config.py rename to paddlespeech/t2s/training/default_config.py diff --git a/parakeet/training/experiment.py b/paddlespeech/t2s/training/experiment.py similarity index 99% rename from parakeet/training/experiment.py rename to paddlespeech/t2s/training/experiment.py index 739fd75d..7a6a7e99 100644 --- a/parakeet/training/experiment.py +++ b/paddlespeech/t2s/training/experiment.py @@ -20,8 +20,8 @@ from paddle import distributed as dist from paddle.io import DistributedBatchSampler from visualdl import LogWriter -from parakeet.utils import checkpoint -from parakeet.utils import mp_tools +from paddlespeech.t2s.utils import checkpoint +from paddlespeech.t2s.utils import mp_tools __all__ = ["ExperimentBase"] diff --git a/parakeet/training/extension.py b/paddlespeech/t2s/training/extension.py similarity index 100% rename from parakeet/training/extension.py rename to paddlespeech/t2s/training/extension.py diff --git a/parakeet/training/extensions/__init__.py b/paddlespeech/t2s/training/extensions/__init__.py similarity index 100% rename from parakeet/training/extensions/__init__.py rename to paddlespeech/t2s/training/extensions/__init__.py diff --git a/parakeet/training/extensions/evaluator.py b/paddlespeech/t2s/training/extensions/evaluator.py similarity index 91% rename from parakeet/training/extensions/evaluator.py rename to paddlespeech/t2s/training/extensions/evaluator.py index 47b35272..28249d09 100644 --- a/parakeet/training/extensions/evaluator.py +++ b/paddlespeech/t2s/training/extensions/evaluator.py @@ -17,10 +17,10 @@ import paddle from paddle.io import DataLoader from paddle.nn import Layer -from parakeet.training import extension -from parakeet.training.reporter import DictSummary -from parakeet.training.reporter import report -from parakeet.training.reporter import scope +from paddlespeech.t2s.training import extension +from paddlespeech.t2s.training.reporter import DictSummary +from paddlespeech.t2s.training.reporter import report +from paddlespeech.t2s.training.reporter import scope class StandardEvaluator(extension.Extension): diff --git a/parakeet/training/extensions/snapshot.py b/paddlespeech/t2s/training/extensions/snapshot.py similarity index 95% rename from parakeet/training/extensions/snapshot.py rename to paddlespeech/t2s/training/extensions/snapshot.py index 7806dd6e..2e401596 100644 --- a/parakeet/training/extensions/snapshot.py +++ b/paddlespeech/t2s/training/extensions/snapshot.py @@ -21,9 +21,9 @@ from typing import List import jsonlines -from parakeet.training import extension -from parakeet.training.trainer import Trainer -from parakeet.utils.mp_tools import rank_zero_only +from paddlespeech.t2s.training import extension +from paddlespeech.t2s.training.trainer import Trainer +from paddlespeech.t2s.utils.mp_tools import rank_zero_only def load_records(records_fp): diff --git a/parakeet/training/extensions/visualizer.py b/paddlespeech/t2s/training/extensions/visualizer.py similarity index 93% rename from parakeet/training/extensions/visualizer.py rename to paddlespeech/t2s/training/extensions/visualizer.py index bc62c976..748a7c48 100644 --- a/parakeet/training/extensions/visualizer.py +++ b/paddlespeech/t2s/training/extensions/visualizer.py @@ -13,8 +13,8 @@ # limitations under the License. from visualdl import LogWriter -from parakeet.training import extension -from parakeet.training.trainer import Trainer +from paddlespeech.t2s.training import extension +from paddlespeech.t2s.training.trainer import Trainer class VisualDL(extension.Extension): diff --git a/parakeet/training/optimizer.py b/paddlespeech/t2s/training/optimizer.py similarity index 100% rename from parakeet/training/optimizer.py rename to paddlespeech/t2s/training/optimizer.py diff --git a/parakeet/training/reporter.py b/paddlespeech/t2s/training/reporter.py similarity index 100% rename from parakeet/training/reporter.py rename to paddlespeech/t2s/training/reporter.py diff --git a/parakeet/training/seeding.py b/paddlespeech/t2s/training/seeding.py similarity index 100% rename from parakeet/training/seeding.py rename to paddlespeech/t2s/training/seeding.py diff --git a/parakeet/training/trainer.py b/paddlespeech/t2s/training/trainer.py similarity index 94% rename from parakeet/training/trainer.py rename to paddlespeech/t2s/training/trainer.py index dd4862f5..9a32bca8 100644 --- a/parakeet/training/trainer.py +++ b/paddlespeech/t2s/training/trainer.py @@ -21,13 +21,13 @@ from typing import Union import six -from parakeet.training.extension import Extension -from parakeet.training.extension import PRIORITY_READER -from parakeet.training.reporter import scope -from parakeet.training.trigger import get_trigger -from parakeet.training.triggers.limit_trigger import LimitTrigger -from parakeet.training.updater import UpdaterBase -from parakeet.utils import profiler +from paddlespeech.t2s.training.extension import Extension +from paddlespeech.t2s.training.extension import PRIORITY_READER +from paddlespeech.t2s.training.reporter import scope +from paddlespeech.t2s.training.trigger import get_trigger +from paddlespeech.t2s.training.triggers.limit_trigger import LimitTrigger +from paddlespeech.t2s.training.updater import UpdaterBase +from paddlespeech.t2s.utils import profiler class _ExtensionEntry(object): diff --git a/parakeet/training/trigger.py b/paddlespeech/t2s/training/trigger.py similarity index 91% rename from parakeet/training/trigger.py rename to paddlespeech/t2s/training/trigger.py index aae1d544..28995623 100644 --- a/parakeet/training/trigger.py +++ b/paddlespeech/t2s/training/trigger.py @@ -11,7 +11,7 @@ # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. -from parakeet.training.triggers.interval_trigger import IntervalTrigger +from paddlespeech.t2s.training.triggers.interval_trigger import IntervalTrigger def never_fail_trigger(trainer): diff --git a/parakeet/training/triggers/__init__.py b/paddlespeech/t2s/training/triggers/__init__.py similarity index 100% rename from parakeet/training/triggers/__init__.py rename to paddlespeech/t2s/training/triggers/__init__.py diff --git a/parakeet/training/triggers/interval_trigger.py b/paddlespeech/t2s/training/triggers/interval_trigger.py similarity index 100% rename from parakeet/training/triggers/interval_trigger.py rename to paddlespeech/t2s/training/triggers/interval_trigger.py diff --git a/parakeet/training/triggers/limit_trigger.py b/paddlespeech/t2s/training/triggers/limit_trigger.py similarity index 100% rename from parakeet/training/triggers/limit_trigger.py rename to paddlespeech/t2s/training/triggers/limit_trigger.py diff --git a/parakeet/training/triggers/time_trigger.py b/paddlespeech/t2s/training/triggers/time_trigger.py similarity index 100% rename from parakeet/training/triggers/time_trigger.py rename to paddlespeech/t2s/training/triggers/time_trigger.py diff --git a/parakeet/training/updater.py b/paddlespeech/t2s/training/updater.py similarity index 100% rename from parakeet/training/updater.py rename to paddlespeech/t2s/training/updater.py diff --git a/parakeet/training/updaters/__init__.py b/paddlespeech/t2s/training/updaters/__init__.py similarity index 100% rename from parakeet/training/updaters/__init__.py rename to paddlespeech/t2s/training/updaters/__init__.py diff --git a/parakeet/training/updaters/standard_updater.py b/paddlespeech/t2s/training/updaters/standard_updater.py similarity index 97% rename from parakeet/training/updaters/standard_updater.py rename to paddlespeech/t2s/training/updaters/standard_updater.py index ea3ec3fa..2d00d196 100644 --- a/parakeet/training/updaters/standard_updater.py +++ b/paddlespeech/t2s/training/updaters/standard_updater.py @@ -23,9 +23,9 @@ from paddle.nn import Layer from paddle.optimizer import Optimizer from timer import timer -from parakeet.training.reporter import report -from parakeet.training.updater import UpdaterBase -from parakeet.training.updater import UpdaterState +from paddlespeech.t2s.training.reporter import report +from paddlespeech.t2s.training.updater import UpdaterBase +from paddlespeech.t2s.training.updater import UpdaterState class StandardUpdater(UpdaterBase): diff --git a/parakeet/utils/__init__.py b/paddlespeech/t2s/utils/__init__.py similarity index 100% rename from parakeet/utils/__init__.py rename to paddlespeech/t2s/utils/__init__.py diff --git a/parakeet/utils/checkpoint.py b/paddlespeech/t2s/utils/checkpoint.py similarity index 99% rename from parakeet/utils/checkpoint.py rename to paddlespeech/t2s/utils/checkpoint.py index 8df791bb..1e222c50 100644 --- a/parakeet/utils/checkpoint.py +++ b/paddlespeech/t2s/utils/checkpoint.py @@ -16,7 +16,7 @@ import os import paddle from paddle import distributed as dist -from parakeet.utils import mp_tools +from paddlespeech.t2s.utils import mp_tools __all__ = ["load_parameters", "save_parameters"] diff --git a/parakeet/utils/display.py b/paddlespeech/t2s/utils/display.py similarity index 100% rename from parakeet/utils/display.py rename to paddlespeech/t2s/utils/display.py diff --git a/parakeet/utils/error_rate.py b/paddlespeech/t2s/utils/error_rate.py similarity index 100% rename from parakeet/utils/error_rate.py rename to paddlespeech/t2s/utils/error_rate.py diff --git a/parakeet/utils/h5_utils.py b/paddlespeech/t2s/utils/h5_utils.py similarity index 100% rename from parakeet/utils/h5_utils.py rename to paddlespeech/t2s/utils/h5_utils.py diff --git a/parakeet/utils/internals.py b/paddlespeech/t2s/utils/internals.py similarity index 100% rename from parakeet/utils/internals.py rename to paddlespeech/t2s/utils/internals.py diff --git a/parakeet/utils/layer_tools.py b/paddlespeech/t2s/utils/layer_tools.py similarity index 100% rename from parakeet/utils/layer_tools.py rename to paddlespeech/t2s/utils/layer_tools.py diff --git a/parakeet/utils/mp_tools.py b/paddlespeech/t2s/utils/mp_tools.py similarity index 100% rename from parakeet/utils/mp_tools.py rename to paddlespeech/t2s/utils/mp_tools.py diff --git a/parakeet/utils/profile.py b/paddlespeech/t2s/utils/profile.py similarity index 100% rename from parakeet/utils/profile.py rename to paddlespeech/t2s/utils/profile.py diff --git a/parakeet/utils/profiler.py b/paddlespeech/t2s/utils/profiler.py similarity index 100% rename from parakeet/utils/profiler.py rename to paddlespeech/t2s/utils/profiler.py diff --git a/parakeet/utils/scheduler.py b/paddlespeech/t2s/utils/scheduler.py similarity index 100% rename from parakeet/utils/scheduler.py rename to paddlespeech/t2s/utils/scheduler.py diff --git a/parakeet/utils/timeline.py b/paddlespeech/t2s/utils/timeline.py similarity index 100% rename from parakeet/utils/timeline.py rename to paddlespeech/t2s/utils/timeline.py diff --git a/text_processing/.gitignore b/paddlespeech/text/.gitignore similarity index 100% rename from text_processing/.gitignore rename to paddlespeech/text/.gitignore diff --git a/text_processing/README.md b/paddlespeech/text/README.md similarity index 100% rename from text_processing/README.md rename to paddlespeech/text/README.md diff --git a/text_processing/examples/punctuation_restoration/chinese/README.md b/paddlespeech/text/examples/punctuation_restoration/chinese/README.md similarity index 100% rename from text_processing/examples/punctuation_restoration/chinese/README.md rename to paddlespeech/text/examples/punctuation_restoration/chinese/README.md diff --git a/text_processing/examples/punctuation_restoration/chinese/conf/blstm.yaml b/paddlespeech/text/examples/punctuation_restoration/chinese/conf/blstm.yaml similarity index 100% rename from text_processing/examples/punctuation_restoration/chinese/conf/blstm.yaml rename to paddlespeech/text/examples/punctuation_restoration/chinese/conf/blstm.yaml diff --git a/text_processing/examples/punctuation_restoration/chinese/conf/data_conf/chinese.yaml b/paddlespeech/text/examples/punctuation_restoration/chinese/conf/data_conf/chinese.yaml similarity index 100% rename from text_processing/examples/punctuation_restoration/chinese/conf/data_conf/chinese.yaml rename to paddlespeech/text/examples/punctuation_restoration/chinese/conf/data_conf/chinese.yaml diff --git a/text_processing/examples/punctuation_restoration/chinese/conf/train_conf/bertBLSTM_zh.yaml b/paddlespeech/text/examples/punctuation_restoration/chinese/conf/train_conf/bertBLSTM_zh.yaml similarity index 100% rename from text_processing/examples/punctuation_restoration/chinese/conf/train_conf/bertBLSTM_zh.yaml rename to paddlespeech/text/examples/punctuation_restoration/chinese/conf/train_conf/bertBLSTM_zh.yaml diff --git a/text_processing/examples/punctuation_restoration/chinese/conf/train_conf/bertLinear_zh.yaml b/paddlespeech/text/examples/punctuation_restoration/chinese/conf/train_conf/bertLinear_zh.yaml similarity index 100% rename from text_processing/examples/punctuation_restoration/chinese/conf/train_conf/bertLinear_zh.yaml rename to paddlespeech/text/examples/punctuation_restoration/chinese/conf/train_conf/bertLinear_zh.yaml diff --git a/text_processing/examples/punctuation_restoration/chinese/local/avg.sh b/paddlespeech/text/examples/punctuation_restoration/chinese/local/avg.sh similarity index 100% rename from text_processing/examples/punctuation_restoration/chinese/local/avg.sh rename to paddlespeech/text/examples/punctuation_restoration/chinese/local/avg.sh diff --git a/text_processing/examples/punctuation_restoration/chinese/local/data.sh b/paddlespeech/text/examples/punctuation_restoration/chinese/local/data.sh similarity index 100% rename from text_processing/examples/punctuation_restoration/chinese/local/data.sh rename to paddlespeech/text/examples/punctuation_restoration/chinese/local/data.sh diff --git a/text_processing/examples/punctuation_restoration/chinese/local/test.sh b/paddlespeech/text/examples/punctuation_restoration/chinese/local/test.sh similarity index 100% rename from text_processing/examples/punctuation_restoration/chinese/local/test.sh rename to paddlespeech/text/examples/punctuation_restoration/chinese/local/test.sh diff --git a/text_processing/examples/punctuation_restoration/chinese/local/train.sh b/paddlespeech/text/examples/punctuation_restoration/chinese/local/train.sh similarity index 100% rename from text_processing/examples/punctuation_restoration/chinese/local/train.sh rename to paddlespeech/text/examples/punctuation_restoration/chinese/local/train.sh diff --git a/text_processing/examples/punctuation_restoration/chinese/path.sh b/paddlespeech/text/examples/punctuation_restoration/chinese/path.sh similarity index 100% rename from text_processing/examples/punctuation_restoration/chinese/path.sh rename to paddlespeech/text/examples/punctuation_restoration/chinese/path.sh diff --git a/text_processing/examples/punctuation_restoration/chinese/run.sh b/paddlespeech/text/examples/punctuation_restoration/chinese/run.sh similarity index 100% rename from text_processing/examples/punctuation_restoration/chinese/run.sh rename to paddlespeech/text/examples/punctuation_restoration/chinese/run.sh diff --git a/text_processing/examples/punctuation_restoration/english/README.md b/paddlespeech/text/examples/punctuation_restoration/english/README.md similarity index 100% rename from text_processing/examples/punctuation_restoration/english/README.md rename to paddlespeech/text/examples/punctuation_restoration/english/README.md diff --git a/text_processing/examples/punctuation_restoration/english/conf/data_conf/english.yaml b/paddlespeech/text/examples/punctuation_restoration/english/conf/data_conf/english.yaml similarity index 100% rename from text_processing/examples/punctuation_restoration/english/conf/data_conf/english.yaml rename to paddlespeech/text/examples/punctuation_restoration/english/conf/data_conf/english.yaml diff --git a/text_processing/examples/punctuation_restoration/english/conf/train_conf/bertBLSTM_base_en.yaml b/paddlespeech/text/examples/punctuation_restoration/english/conf/train_conf/bertBLSTM_base_en.yaml similarity index 100% rename from text_processing/examples/punctuation_restoration/english/conf/train_conf/bertBLSTM_base_en.yaml rename to paddlespeech/text/examples/punctuation_restoration/english/conf/train_conf/bertBLSTM_base_en.yaml diff --git a/text_processing/examples/punctuation_restoration/english/conf/train_conf/bertLinear_en.yaml b/paddlespeech/text/examples/punctuation_restoration/english/conf/train_conf/bertLinear_en.yaml similarity index 100% rename from text_processing/examples/punctuation_restoration/english/conf/train_conf/bertLinear_en.yaml rename to paddlespeech/text/examples/punctuation_restoration/english/conf/train_conf/bertLinear_en.yaml diff --git a/text_processing/examples/punctuation_restoration/english/local/avg.sh b/paddlespeech/text/examples/punctuation_restoration/english/local/avg.sh similarity index 100% rename from text_processing/examples/punctuation_restoration/english/local/avg.sh rename to paddlespeech/text/examples/punctuation_restoration/english/local/avg.sh diff --git a/text_processing/examples/punctuation_restoration/english/local/data.sh b/paddlespeech/text/examples/punctuation_restoration/english/local/data.sh similarity index 100% rename from text_processing/examples/punctuation_restoration/english/local/data.sh rename to paddlespeech/text/examples/punctuation_restoration/english/local/data.sh diff --git a/text_processing/examples/punctuation_restoration/english/local/test.sh b/paddlespeech/text/examples/punctuation_restoration/english/local/test.sh similarity index 100% rename from text_processing/examples/punctuation_restoration/english/local/test.sh rename to paddlespeech/text/examples/punctuation_restoration/english/local/test.sh diff --git a/text_processing/examples/punctuation_restoration/english/local/train.sh b/paddlespeech/text/examples/punctuation_restoration/english/local/train.sh similarity index 100% rename from text_processing/examples/punctuation_restoration/english/local/train.sh rename to paddlespeech/text/examples/punctuation_restoration/english/local/train.sh diff --git a/text_processing/examples/punctuation_restoration/english/path.sh b/paddlespeech/text/examples/punctuation_restoration/english/path.sh similarity index 100% rename from text_processing/examples/punctuation_restoration/english/path.sh rename to paddlespeech/text/examples/punctuation_restoration/english/path.sh diff --git a/text_processing/examples/punctuation_restoration/english/run.sh b/paddlespeech/text/examples/punctuation_restoration/english/run.sh similarity index 100% rename from text_processing/examples/punctuation_restoration/english/run.sh rename to paddlespeech/text/examples/punctuation_restoration/english/run.sh diff --git a/text_processing/requirements.txt b/paddlespeech/text/requirements.txt similarity index 100% rename from text_processing/requirements.txt rename to paddlespeech/text/requirements.txt diff --git a/text_processing/speechtask/punctuation_restoration/bin/avg_model.py b/paddlespeech/text/speechtask/punctuation_restoration/bin/avg_model.py similarity index 100% rename from text_processing/speechtask/punctuation_restoration/bin/avg_model.py rename to paddlespeech/text/speechtask/punctuation_restoration/bin/avg_model.py diff --git a/text_processing/speechtask/punctuation_restoration/bin/pre_data.py b/paddlespeech/text/speechtask/punctuation_restoration/bin/pre_data.py similarity index 100% rename from text_processing/speechtask/punctuation_restoration/bin/pre_data.py rename to paddlespeech/text/speechtask/punctuation_restoration/bin/pre_data.py diff --git a/text_processing/speechtask/punctuation_restoration/bin/test.py b/paddlespeech/text/speechtask/punctuation_restoration/bin/test.py similarity index 100% rename from text_processing/speechtask/punctuation_restoration/bin/test.py rename to paddlespeech/text/speechtask/punctuation_restoration/bin/test.py diff --git a/text_processing/speechtask/punctuation_restoration/bin/train.py b/paddlespeech/text/speechtask/punctuation_restoration/bin/train.py similarity index 100% rename from text_processing/speechtask/punctuation_restoration/bin/train.py rename to paddlespeech/text/speechtask/punctuation_restoration/bin/train.py diff --git a/text_processing/speechtask/punctuation_restoration/io/__init__.py b/paddlespeech/text/speechtask/punctuation_restoration/io/__init__.py similarity index 100% rename from text_processing/speechtask/punctuation_restoration/io/__init__.py rename to paddlespeech/text/speechtask/punctuation_restoration/io/__init__.py diff --git a/text_processing/speechtask/punctuation_restoration/io/collator.py b/paddlespeech/text/speechtask/punctuation_restoration/io/collator.py similarity index 100% rename from text_processing/speechtask/punctuation_restoration/io/collator.py rename to paddlespeech/text/speechtask/punctuation_restoration/io/collator.py diff --git a/text_processing/speechtask/punctuation_restoration/io/common.py b/paddlespeech/text/speechtask/punctuation_restoration/io/common.py similarity index 100% rename from text_processing/speechtask/punctuation_restoration/io/common.py rename to paddlespeech/text/speechtask/punctuation_restoration/io/common.py diff --git a/text_processing/speechtask/punctuation_restoration/io/dataset.py b/paddlespeech/text/speechtask/punctuation_restoration/io/dataset.py similarity index 100% rename from text_processing/speechtask/punctuation_restoration/io/dataset.py rename to paddlespeech/text/speechtask/punctuation_restoration/io/dataset.py diff --git a/text_processing/speechtask/punctuation_restoration/model/BertBLSTM.py b/paddlespeech/text/speechtask/punctuation_restoration/model/BertBLSTM.py similarity index 100% rename from text_processing/speechtask/punctuation_restoration/model/BertBLSTM.py rename to paddlespeech/text/speechtask/punctuation_restoration/model/BertBLSTM.py diff --git a/text_processing/speechtask/punctuation_restoration/model/BertLinear.py b/paddlespeech/text/speechtask/punctuation_restoration/model/BertLinear.py similarity index 100% rename from text_processing/speechtask/punctuation_restoration/model/BertLinear.py rename to paddlespeech/text/speechtask/punctuation_restoration/model/BertLinear.py diff --git a/text_processing/speechtask/punctuation_restoration/model/blstm.py b/paddlespeech/text/speechtask/punctuation_restoration/model/blstm.py similarity index 100% rename from text_processing/speechtask/punctuation_restoration/model/blstm.py rename to paddlespeech/text/speechtask/punctuation_restoration/model/blstm.py diff --git a/text_processing/speechtask/punctuation_restoration/model/lstm.py b/paddlespeech/text/speechtask/punctuation_restoration/model/lstm.py similarity index 100% rename from text_processing/speechtask/punctuation_restoration/model/lstm.py rename to paddlespeech/text/speechtask/punctuation_restoration/model/lstm.py diff --git a/text_processing/speechtask/punctuation_restoration/modules/__init__.py b/paddlespeech/text/speechtask/punctuation_restoration/modules/__init__.py similarity index 100% rename from text_processing/speechtask/punctuation_restoration/modules/__init__.py rename to paddlespeech/text/speechtask/punctuation_restoration/modules/__init__.py diff --git a/text_processing/speechtask/punctuation_restoration/modules/activation.py b/paddlespeech/text/speechtask/punctuation_restoration/modules/activation.py similarity index 100% rename from text_processing/speechtask/punctuation_restoration/modules/activation.py rename to paddlespeech/text/speechtask/punctuation_restoration/modules/activation.py diff --git a/text_processing/speechtask/punctuation_restoration/modules/attention.py b/paddlespeech/text/speechtask/punctuation_restoration/modules/attention.py similarity index 100% rename from text_processing/speechtask/punctuation_restoration/modules/attention.py rename to paddlespeech/text/speechtask/punctuation_restoration/modules/attention.py diff --git a/text_processing/speechtask/punctuation_restoration/modules/crf.py b/paddlespeech/text/speechtask/punctuation_restoration/modules/crf.py similarity index 100% rename from text_processing/speechtask/punctuation_restoration/modules/crf.py rename to paddlespeech/text/speechtask/punctuation_restoration/modules/crf.py diff --git a/text_processing/speechtask/punctuation_restoration/training/__init__.py b/paddlespeech/text/speechtask/punctuation_restoration/training/__init__.py similarity index 100% rename from text_processing/speechtask/punctuation_restoration/training/__init__.py rename to paddlespeech/text/speechtask/punctuation_restoration/training/__init__.py diff --git a/text_processing/speechtask/punctuation_restoration/training/loss.py b/paddlespeech/text/speechtask/punctuation_restoration/training/loss.py similarity index 100% rename from text_processing/speechtask/punctuation_restoration/training/loss.py rename to paddlespeech/text/speechtask/punctuation_restoration/training/loss.py diff --git a/text_processing/speechtask/punctuation_restoration/training/trainer.py b/paddlespeech/text/speechtask/punctuation_restoration/training/trainer.py similarity index 100% rename from text_processing/speechtask/punctuation_restoration/training/trainer.py rename to paddlespeech/text/speechtask/punctuation_restoration/training/trainer.py diff --git a/text_processing/speechtask/punctuation_restoration/utils/__init__.py b/paddlespeech/text/speechtask/punctuation_restoration/utils/__init__.py similarity index 100% rename from text_processing/speechtask/punctuation_restoration/utils/__init__.py rename to paddlespeech/text/speechtask/punctuation_restoration/utils/__init__.py diff --git a/text_processing/speechtask/punctuation_restoration/utils/checkpoint.py b/paddlespeech/text/speechtask/punctuation_restoration/utils/checkpoint.py similarity index 100% rename from text_processing/speechtask/punctuation_restoration/utils/checkpoint.py rename to paddlespeech/text/speechtask/punctuation_restoration/utils/checkpoint.py diff --git a/text_processing/speechtask/punctuation_restoration/utils/default_parser.py b/paddlespeech/text/speechtask/punctuation_restoration/utils/default_parser.py similarity index 93% rename from text_processing/speechtask/punctuation_restoration/utils/default_parser.py rename to paddlespeech/text/speechtask/punctuation_restoration/utils/default_parser.py index b83d989d..405b29a2 100644 --- a/text_processing/speechtask/punctuation_restoration/utils/default_parser.py +++ b/paddlespeech/text/speechtask/punctuation_restoration/utils/default_parser.py @@ -15,10 +15,10 @@ import argparse def default_argument_parser(): - r"""A simple yet genral argument parser for experiments with parakeet. + r"""A simple yet genral argument parser for experiments with t2s. - This is used in examples with parakeet. And it is intended to be used by - other experiments with parakeet. It requires a minimal set of command line + This is used in examples with t2s. And it is intended to be used by + other experiments with t2s. It requires a minimal set of command line arguments to start a training script. The ``--config`` and ``--opts`` are used for overwrite the deault @@ -35,7 +35,7 @@ def default_argument_parser(): See Also -------- - parakeet.training.experiment + paddlespeech.t2s.training.experiment Returns ------- argparse.ArgumentParser diff --git a/text_processing/speechtask/punctuation_restoration/utils/layer_tools.py b/paddlespeech/text/speechtask/punctuation_restoration/utils/layer_tools.py similarity index 100% rename from text_processing/speechtask/punctuation_restoration/utils/layer_tools.py rename to paddlespeech/text/speechtask/punctuation_restoration/utils/layer_tools.py diff --git a/text_processing/speechtask/punctuation_restoration/utils/mp_tools.py b/paddlespeech/text/speechtask/punctuation_restoration/utils/mp_tools.py similarity index 100% rename from text_processing/speechtask/punctuation_restoration/utils/mp_tools.py rename to paddlespeech/text/speechtask/punctuation_restoration/utils/mp_tools.py diff --git a/text_processing/speechtask/punctuation_restoration/utils/punct_pre.py b/paddlespeech/text/speechtask/punctuation_restoration/utils/punct_pre.py similarity index 100% rename from text_processing/speechtask/punctuation_restoration/utils/punct_pre.py rename to paddlespeech/text/speechtask/punctuation_restoration/utils/punct_pre.py diff --git a/text_processing/speechtask/punctuation_restoration/utils/utility.py b/paddlespeech/text/speechtask/punctuation_restoration/utils/utility.py similarity index 100% rename from text_processing/speechtask/punctuation_restoration/utils/utility.py rename to paddlespeech/text/speechtask/punctuation_restoration/utils/utility.py diff --git a/parakeet/frontend/normalizer/__init__.py b/parakeet/frontend/normalizer/__init__.py deleted file mode 100644 index d1f2bfc5..00000000 --- a/parakeet/frontend/normalizer/__init__.py +++ /dev/null @@ -1,15 +0,0 @@ -# Copyright (c) 2020 PaddlePaddle Authors. All Rights Reserved. -# -# Licensed under the Apache License, Version 2.0 (the "License"); -# you may not use this file except in compliance with the License. -# You may obtain a copy of the License at -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. -from parakeet.frontend.normalizer.normalizer import * -from parakeet.frontend.normalizer.numbers import * diff --git a/parakeet/frontend/zh_normalization/__init__.py b/parakeet/frontend/zh_normalization/__init__.py deleted file mode 100644 index 1e494080..00000000 --- a/parakeet/frontend/zh_normalization/__init__.py +++ /dev/null @@ -1,14 +0,0 @@ -# Copyright (c) 2020 PaddlePaddle Authors. All Rights Reserved. -# -# Licensed under the Apache License, Version 2.0 (the "License"); -# you may not use this file except in compliance with the License. -# You may obtain a copy of the License at -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. -from parakeet.frontend.zh_normalization.text_normlization import * diff --git a/setup.py b/setup.py index a2e4c031..95044e74 100644 --- a/setup.py +++ b/setup.py @@ -80,7 +80,7 @@ def _post_install(install_lib_dir): print("autolog install.") # ctcdecoder - ctcdecoder_dir = HERE / 'deepspeech/decoders/ctcdecoder/swig' + ctcdecoder_dir = HERE / 'paddlespeech/s2t/decoders/ctcdecoder/swig' with pushd(ctcdecoder_dir): check_call("bash -e setup.sh") print("ctcdecoder install.") diff --git a/setup.sh b/setup.sh index aefdab98..0f04226d 100644 --- a/setup.sh +++ b/setup.sh @@ -69,7 +69,7 @@ fi # install decoders python3 -c "import pkg_resources; pkg_resources.require(\"swig_decoders==1.1\")" if [ $? != 0 ]; then - cd deepspeech/decoders/ctcdecoder/swig > /dev/null + cd paddlespeech/s2t/decoders/ctcdecoder/swig > /dev/null bash setup.sh cd - > /dev/null fi diff --git a/tests/benchmark/pwgan/README.md b/tests/benchmark/pwgan/README.md index 369f4b74..bfc2bc07 100644 --- a/tests/benchmark/pwgan/README.md +++ b/tests/benchmark/pwgan/README.md @@ -5,7 +5,7 @@ 即可运行. 执行逻辑： 1. cd 到 ../../../ (也就是 Deepspeech 目录) -2. 安装 parakeet 所需依赖 +2. 安装 paddlespeech/t2s 所需依赖 3. 从 bos 下载数据集并解压缩 4. 预处理数据集为训练 pwg 所需格式，保存到 Deepspeech/dump 文件夹底下 5. 按照不同的参数执行 run_benchmark.sh 脚本 diff --git a/tests/benchmark/pwgan/run_all.sh b/tests/benchmark/pwgan/run_all.sh index e4bb17f8..9d722aaf 100755 --- a/tests/benchmark/pwgan/run_all.sh +++ b/tests/benchmark/pwgan/run_all.sh @@ -25,11 +25,11 @@ fi # 数据预处理 if [ ${stage} -le 2 ] && [ ${stop_stage} -ge 2 ]; then - python3 parakeet/exps/gan_vocoder/preprocess.py --rootdir=BZNSYP/ --dumpdir=dump --num-cpu=20 --cut-sil=True --dur-file=durations.txt --config=examples/csmsc/voc1/conf/default.yaml + python3 paddlespeech/t2s/exps/gan_vocoder/preprocess.py --rootdir=BZNSYP/ --dumpdir=dump --num-cpu=20 --cut-sil=True --dur-file=durations.txt --config=examples/csmsc/voc1/conf/default.yaml python3 utils/compute_statistics.py --metadata=dump/train/raw/metadata.jsonl --field-name="feats" - python3 parakeet/exps/gan_vocoder/normalize.py --metadata=dump/train/raw/metadata.jsonl --dumpdir=dump/train/norm --stats=dump/train/feats_stats.npy - python3 parakeet/exps/gan_vocoder/normalize.py --metadata=dump/dev/raw/metadata.jsonl --dumpdir=dump/dev/norm --stats=dump/train/feats_stats.npy - python3 parakeet/exps/gan_vocoder/normalize.py --metadata=dump/test/raw/metadata.jsonl --dumpdir=dump/test/norm --stats=dump/train/feats_stats.npy + python3 paddlespeech/t2s/exps/gan_vocoder/normalize.py --metadata=dump/train/raw/metadata.jsonl --dumpdir=dump/train/norm --stats=dump/train/feats_stats.npy + python3 paddlespeech/t2s/exps/gan_vocoder/normalize.py --metadata=dump/dev/raw/metadata.jsonl --dumpdir=dump/dev/norm --stats=dump/train/feats_stats.npy + python3 paddlespeech/t2s/exps/gan_vocoder/normalize.py --metadata=dump/test/raw/metadata.jsonl --dumpdir=dump/test/norm --stats=dump/train/feats_stats.npy fi # 3 批量运行（如不方便批量，1，2需放到单个模型中） if [ ${stage} -le 3 ] && [ ${stop_stage} -ge 3 ]; then diff --git a/tests/benchmark/pwgan/run_benchmark.sh b/tests/benchmark/pwgan/run_benchmark.sh index e961e442..be5733da 100755 --- a/tests/benchmark/pwgan/run_benchmark.sh +++ b/tests/benchmark/pwgan/run_benchmark.sh @@ -29,8 +29,8 @@ function _train(){ --run-benchmark=true" case ${run_mode} in - sp) train_cmd="python3 parakeet/exps/gan_vocoder/parallelwave_gan/train.py --nprocs=1 ${train_cmd}" ;; - mp) train_cmd="python3 parakeet/exps/gan_vocoder/parallelwave_gan/train.py --nprocs=8 ${train_cmd}" + sp) train_cmd="python3 paddlespeech/t2s/exps/gan_vocoder/parallelwave_gan/train.py --nprocs=1 ${train_cmd}" ;; + mp) train_cmd="python3 paddlespeech/t2s/exps/gan_vocoder/parallelwave_gan/train.py --nprocs=8 ${train_cmd}" log_parse_file="mylog/workerlog.0" ;; *) echo "choose run_mode(sp or mp)"; exit 1; esac diff --git a/tests/chains/ds2/ds2_params_lite_train_infer.txt b/tests/chains/ds2/ds2_params_lite_train_infer.txt index 70d54f8b..c1cbfbb9 100644 --- a/tests/chains/ds2/ds2_params_lite_train_infer.txt +++ b/tests/chains/ds2/ds2_params_lite_train_infer.txt @@ -13,7 +13,7 @@ null:null null:null ## trainer:norm_train -norm_train: ../../../deepspeech/exps/deepspeech2/bin/train.py --config conf/deepspeech2.yaml --model_type offline --profiler-options "" --output exp/deepspeech_tiny --seed 0 +norm_train: ../../../paddlespeech/s2t/exps/deepspeech2/bin/train.py --config conf/deepspeech2.yaml --model_type offline --profiler-options "" --output exp/deepspeech_tiny --seed 0 pact_train:null fpgm_train:null distill_train:null @@ -21,13 +21,13 @@ null:null null:null ## ===========================eval_params=========================== -eval: ../../../deepspeech/exps/deepspeech2/bin/test.py --nproc 1 --config conf/deepspeech2.yaml --checkpoint_path exp/deepspeech_tiny/checkpoints/9 --result_file tests/9.rsl --model_type offline +eval: ../../../paddlespeech/s2t/exps/deepspeech2/bin/test.py --nproc 1 --config conf/deepspeech2.yaml --checkpoint_path exp/deepspeech_tiny/checkpoints/9 --result_file tests/9.rsl --model_type offline null:null ## ===========================infer_params=========================== null:null null:null -norm_export: ../../../deepspeech/exps/deepspeech2/bin/export.py --nproc 1 --config conf/deepspeech2.yaml --model_type offline --checkpoint_path exp/deepspeech_tiny/checkpoints/9 --export_path exp/deepspeech_tiny/checkpoints/9.jit +norm_export: ../../../paddlespeech/s2t/exps/deepspeech2/bin/export.py --nproc 1 --config conf/deepspeech2.yaml --model_type offline --checkpoint_path exp/deepspeech_tiny/checkpoints/9 --export_path exp/deepspeech_tiny/checkpoints/9.jit quant_export:null fpgm_export:null distill_export:null diff --git a/tests/chains/ds2/ds2_params_whole_train_infer.txt b/tests/chains/ds2/ds2_params_whole_train_infer.txt index 90ce7d89..bfcb745f 100644 --- a/tests/chains/ds2/ds2_params_whole_train_infer.txt +++ b/tests/chains/ds2/ds2_params_whole_train_infer.txt @@ -13,7 +13,7 @@ null:null null:null ## trainer:norm_train -norm_train: ../../../deepspeech/exps/deepspeech2/bin/train.py --config conf/deepspeech2.yaml --model_type offline --profiler-options "" --output exp/deepspeech_whole --seed 0 +norm_train: ../../../paddlespeech/s2t/exps/deepspeech2/bin/train.py --config conf/deepspeech2.yaml --model_type offline --profiler-options "" --output exp/deepspeech_whole --seed 0 pact_train:null fpgm_train:null distill_train:null @@ -21,13 +21,13 @@ null:null null:null ## ===========================eval_params=========================== -eval: ../../../deepspeech/exps/deepspeech2/bin/test.py --nproc 1 --config conf/deepspeech2.yaml --result_file tests/49.rsl --checkpoint_path exp/deepspeech_whole/checkpoints/49 --model_type offline +eval: ../../../paddlespeech/s2t/exps/deepspeech2/bin/test.py --nproc 1 --config conf/deepspeech2.yaml --result_file tests/49.rsl --checkpoint_path exp/deepspeech_whole/checkpoints/49 --model_type offline null:null ## ===========================infer_params=========================== null:null null:null -norm_export: ../../../deepspeech/exps/deepspeech2/bin/export.py --nproc 1 --config conf/deepspeech2.yaml --model_type offline --checkpoint_path exp/deepspeech_whole/checkpoints/49 --export_path exp/deepspeech_whole/checkpoints/49.jit +norm_export: ../../../paddlespeech/s2t/exps/deepspeech2/bin/export.py --nproc 1 --config conf/deepspeech2.yaml --model_type offline --checkpoint_path exp/deepspeech_whole/checkpoints/49 --export_path exp/deepspeech_whole/checkpoints/49.jit quant_export:null fpgm_export:null distill_export:null diff --git a/tests/chains/speedyspeech/speedyspeech_params_lite_multi_gpu.txt b/tests/chains/speedyspeech/speedyspeech_params_lite_multi_gpu.txt index ad342052..0f64da27 100644 --- a/tests/chains/speedyspeech/speedyspeech_params_lite_multi_gpu.txt +++ b/tests/chains/speedyspeech/speedyspeech_params_lite_multi_gpu.txt @@ -13,7 +13,7 @@ null:null null:null ## trainer:norm_train -norm_train:../../../parakeet/exps/speedyspeech/train.py --train-metadata=train_data/mini_BZNSYP/train/norm/metadata.jsonl --dev-metadata=train_data/mini_BZNSYP/dev/norm/metadata.jsonl --config=../../../examples/csmsc/tts2/conf/default.yaml --batch_size=32 --max_epoch=20 --num_snapshots=10 --output-dir=exp/default --phones-dict=train_data/mini_BZNSYP/phone_id_map.txt --tones-dict=train_data/mini_BZNSYP/tone_id_map.txt --use-relative-path=True +norm_train:../../../paddlespeech/t2s/exps/speedyspeech/train.py --train-metadata=train_data/mini_BZNSYP/train/norm/metadata.jsonl --dev-metadata=train_data/mini_BZNSYP/dev/norm/metadata.jsonl --config=../../../examples/csmsc/tts2/conf/default.yaml --batch_size=32 --max_epoch=20 --num_snapshots=10 --output-dir=exp/default --phones-dict=train_data/mini_BZNSYP/phone_id_map.txt --tones-dict=train_data/mini_BZNSYP/tone_id_map.txt --use-relative-path=True null:null null:null null:null @@ -21,7 +21,7 @@ null:null null:null ## ===========================eval_params=========================== -eval:../../../parakeet/exps/speedyspeech/synthesize_e2e.py --speedyspeech-config=../../../examples/csmsc/tts2/conf/default.yaml --speedyspeech-checkpoint=exp/default/checkpoints/snapshot_iter_20.pdz --speedyspeech-stat=train_data/mini_BZNSYP/train/feats_stats.npy --pwg-config=pretrain_models/pwg_baker_ckpt_0.4/pwg_default.yaml --pwg-checkpoint=pretrain_models/pwg_baker_ckpt_0.4/pwg_snapshot_iter_400000.pdz --pwg-stat=pretrain_models/pwg_baker_ckpt_0.4/pwg_stats.npy --text=../../../parakeet/exps/sentences.txt --output-dir=e2e --inference-dir=inference --device="gpu" --phones-dict=train_data/mini_BZNSYP/phone_id_map.txt --tones-dict=train_data/mini_BZNSYP/tone_id_map.txt +eval:../../../paddlespeech/t2s/exps/speedyspeech/synthesize_e2e.py --speedyspeech-config=../../../examples/csmsc/tts2/conf/default.yaml --speedyspeech-checkpoint=exp/default/checkpoints/snapshot_iter_20.pdz --speedyspeech-stat=train_data/mini_BZNSYP/train/feats_stats.npy --pwg-config=pretrain_models/pwg_baker_ckpt_0.4/pwg_default.yaml --pwg-checkpoint=pretrain_models/pwg_baker_ckpt_0.4/pwg_snapshot_iter_400000.pdz --pwg-stat=pretrain_models/pwg_baker_ckpt_0.4/pwg_stats.npy --text=../../../paddlespeech/t2s/exps/sentences.txt --output-dir=e2e --inference-dir=inference --device="gpu" --phones-dict=train_data/mini_BZNSYP/phone_id_map.txt --tones-dict=train_data/mini_BZNSYP/tone_id_map.txt null:null ## ===========================infer_params=========================== @@ -37,7 +37,7 @@ null:null null:null null:null null:null -inference:../../../parakeet/exps/speedyspeech/inference.py --inference-dir=pretrain_models/speedyspeech_pwg_inference_0.5 --text=../../../parakeet/exps/sentences.txt --output-dir=inference_out --enable-auto-log --phones-dict=pretrain_models/speedyspeech_pwg_inference_0.5/phone_id_map.txt --tones-dict=pretrain_models/speedyspeech_pwg_inference_0.5/tone_id_map.txt +inference:../../../paddlespeech/t2s/exps/speedyspeech/inference.py --inference-dir=pretrain_models/speedyspeech_pwg_inference_0.5 --text=../../../paddlespeech/t2s/exps/sentences.txt --output-dir=inference_out --enable-auto-log --phones-dict=pretrain_models/speedyspeech_pwg_inference_0.5/phone_id_map.txt --tones-dict=pretrain_models/speedyspeech_pwg_inference_0.5/tone_id_map.txt null:null null:null null:null diff --git a/tests/chains/speedyspeech/speedyspeech_params_lite_single_gpu.txt b/tests/chains/speedyspeech/speedyspeech_params_lite_single_gpu.txt index eaad714d..beda4c04 100644 --- a/tests/chains/speedyspeech/speedyspeech_params_lite_single_gpu.txt +++ b/tests/chains/speedyspeech/speedyspeech_params_lite_single_gpu.txt @@ -13,7 +13,7 @@ null:null null:null ## trainer:norm_train -norm_train:../../../parakeet/exps/speedyspeech/train.py --train-metadata=train_data/mini_BZNSYP/train/norm/metadata.jsonl --dev-metadata=train_data/mini_BZNSYP/dev/norm/metadata.jsonl --config=../../../examples/csmsc/tts2/conf/default.yaml --batch_size=32 --max_epoch=10 --num_snapshots=10 --output-dir=exp/default --phones-dict=train_data/mini_BZNSYP/phone_id_map.txt --tones-dict=train_data/mini_BZNSYP/tone_id_map.txt --use-relative-path=True +norm_train:../../../paddlespeech/t2s/exps/speedyspeech/train.py --train-metadata=train_data/mini_BZNSYP/train/norm/metadata.jsonl --dev-metadata=train_data/mini_BZNSYP/dev/norm/metadata.jsonl --config=../../../examples/csmsc/tts2/conf/default.yaml --batch_size=32 --max_epoch=10 --num_snapshots=10 --output-dir=exp/default --phones-dict=train_data/mini_BZNSYP/phone_id_map.txt --tones-dict=train_data/mini_BZNSYP/tone_id_map.txt --use-relative-path=True null:null null:null null:null @@ -21,7 +21,7 @@ null:null null:null ## ===========================eval_params=========================== -eval:../../../parakeet/exps/speedyspeech/synthesize_e2e.py --speedyspeech-config=../../../examples/csmsc/tts2/conf/default.yaml --speedyspeech-checkpoint=exp/default/checkpoints/snapshot_iter_30.pdz --speedyspeech-stat=train_data/mini_BZNSYP/train/feats_stats.npy --pwg-config=pretrain_models/pwg_baker_ckpt_0.4/pwg_default.yaml --pwg-checkpoint=pretrain_models/pwg_baker_ckpt_0.4/pwg_snapshot_iter_400000.pdz --pwg-stat=pretrain_models/pwg_baker_ckpt_0.4/pwg_stats.npy --text=../../../parakeet/exps/sentences.txt --output-dir=e2e --inference-dir=inference --device="gpu" --phones-dict=train_data/mini_BZNSYP/phone_id_map.txt --tones-dict=train_data/mini_BZNSYP/tone_id_map.txt +eval:../../../paddlespeech/t2s/exps/speedyspeech/synthesize_e2e.py --speedyspeech-config=../../../examples/csmsc/tts2/conf/default.yaml --speedyspeech-checkpoint=exp/default/checkpoints/snapshot_iter_30.pdz --speedyspeech-stat=train_data/mini_BZNSYP/train/feats_stats.npy --pwg-config=pretrain_models/pwg_baker_ckpt_0.4/pwg_default.yaml --pwg-checkpoint=pretrain_models/pwg_baker_ckpt_0.4/pwg_snapshot_iter_400000.pdz --pwg-stat=pretrain_models/pwg_baker_ckpt_0.4/pwg_stats.npy --text=../../../paddlespeech/t2s/exps/sentences.txt --output-dir=e2e --inference-dir=inference --device="gpu" --phones-dict=train_data/mini_BZNSYP/phone_id_map.txt --tones-dict=train_data/mini_BZNSYP/tone_id_map.txt null:null ## ===========================infer_params=========================== @@ -37,7 +37,7 @@ null:null null:null null:null null:null -inference:../../../parakeet/exps/speedyspeech/inference.py --inference-dir=pretrain_models/speedyspeech_pwg_inference_0.5 --text=../../../parakeet/exps/sentences.txt --output-dir=inference_out --enable-auto-log --phones-dict=pretrain_models/speedyspeech_pwg_inference_0.5/phone_id_map.txt --tones-dict=pretrain_models/speedyspeech_pwg_inference_0.5/tone_id_map.txt +inference:../../../paddlespeech/t2s/exps/speedyspeech/inference.py --inference-dir=pretrain_models/speedyspeech_pwg_inference_0.5 --text=../../../paddlespeech/t2s/exps/sentences.txt --output-dir=inference_out --enable-auto-log --phones-dict=pretrain_models/speedyspeech_pwg_inference_0.5/phone_id_map.txt --tones-dict=pretrain_models/speedyspeech_pwg_inference_0.5/tone_id_map.txt --use_gpu:True null:null null:null diff --git a/tests/chains/speedyspeech/speedyspeech_params_whole_multi_gpu.txt b/tests/chains/speedyspeech/speedyspeech_params_whole_multi_gpu.txt index 236805fc..ecdbf76d 100644 --- a/tests/chains/speedyspeech/speedyspeech_params_whole_multi_gpu.txt +++ b/tests/chains/speedyspeech/speedyspeech_params_whole_multi_gpu.txt @@ -13,7 +13,7 @@ null:null null:null ## trainer:norm_train -norm_train:../../../parakeet/exps/speedyspeech/train.py --train-metadata=train_data/processed_BZNSYP/train/norm/metadata.jsonl --dev-metadata=train_data/processed_BZNSYP/dev/norm/metadata.jsonl --config=../../../examples/csmsc/tts2/conf/default.yaml --output-dir=exp/whole --phones-dict=train_data/processed_BZNSYP/phone_id_map.txt --tones-dict=train_data/processed_BZNSYP/tone_id_map.txt --use-relative-path=True +norm_train:../../../paddlespeech/t2s/exps/speedyspeech/train.py --train-metadata=train_data/processed_BZNSYP/train/norm/metadata.jsonl --dev-metadata=train_data/processed_BZNSYP/dev/norm/metadata.jsonl --config=../../../examples/csmsc/tts2/conf/default.yaml --output-dir=exp/whole --phones-dict=train_data/processed_BZNSYP/phone_id_map.txt --tones-dict=train_data/processed_BZNSYP/tone_id_map.txt --use-relative-path=True null:null null:null null:null @@ -21,7 +21,7 @@ null:null null:null ## ===========================eval_params=========================== -eval:../../../parakeet/exps/speedyspeech/synthesize_e2e.py --speedyspeech-config=pretrain_models/speedyspeech_nosil_baker_ckpt_0.5/default.yaml --speedyspeech-checkpoint=pretrain_models/speedyspeech_nosil_baker_ckpt_0.5/snapshot_iter_11400.pdz --speedyspeech-stat=pretrain_models/speedyspeech_nosil_baker_ckpt_0.5/feats_stats.npy --pwg-config=pretrain_models/pwg_baker_ckpt_0.4/pwg_default.yaml --pwg-checkpoint=pretrain_models/pwg_baker_ckpt_0.4/pwg_snapshot_iter_400000.pdz --pwg-stat=pretrain_models/pwg_baker_ckpt_0.4/pwg_stats.npy --text=../../../parakeet/exps/sentences.txt --output-dir=e2e --inference-dir=inference --device="gpu" --phones-dict=pretrain_models/speedyspeech_nosil_baker_ckpt_0.5/phone_id_map.txt --tones-dict=pretrain_models/speedyspeech_nosil_baker_ckpt_0.5/tone_id_map.txt +eval:../../../paddlespeech/t2s/exps/speedyspeech/synthesize_e2e.py --speedyspeech-config=pretrain_models/speedyspeech_nosil_baker_ckpt_0.5/default.yaml --speedyspeech-checkpoint=pretrain_models/speedyspeech_nosil_baker_ckpt_0.5/snapshot_iter_11400.pdz --speedyspeech-stat=pretrain_models/speedyspeech_nosil_baker_ckpt_0.5/feats_stats.npy --pwg-config=pretrain_models/pwg_baker_ckpt_0.4/pwg_default.yaml --pwg-checkpoint=pretrain_models/pwg_baker_ckpt_0.4/pwg_snapshot_iter_400000.pdz --pwg-stat=pretrain_models/pwg_baker_ckpt_0.4/pwg_stats.npy --text=../../../paddlespeech/t2s/exps/sentences.txt --output-dir=e2e --inference-dir=inference --device="gpu" --phones-dict=pretrain_models/speedyspeech_nosil_baker_ckpt_0.5/phone_id_map.txt --tones-dict=pretrain_models/speedyspeech_nosil_baker_ckpt_0.5/tone_id_map.txt null:null ## ===========================infer_params=========================== @@ -37,7 +37,7 @@ null:null null:null null:null null:null -inference:../../../parakeet/exps/speedyspeech/inference.py --inference-dir=pretrain_models/speedyspeech_pwg_inference_0.5 --text=../../../parakeet/exps/sentences.txt --output-dir=inference_out --enable-auto-log --phones-dict=pretrain_models/speedyspeech_pwg_inference_0.5/phone_id_map.txt --tones-dict=pretrain_models/speedyspeech_pwg_inference_0.5/tone_id_map.txt +inference:../../../paddlespeech/t2s/exps/speedyspeech/inference.py --inference-dir=pretrain_models/speedyspeech_pwg_inference_0.5 --text=../../../paddlespeech/t2s/exps/sentences.txt --output-dir=inference_out --enable-auto-log --phones-dict=pretrain_models/speedyspeech_pwg_inference_0.5/phone_id_map.txt --tones-dict=pretrain_models/speedyspeech_pwg_inference_0.5/tone_id_map.txt null:null null:null null:null diff --git a/tests/chains/speedyspeech/speedyspeech_params_whole_single_gpu.txt b/tests/chains/speedyspeech/speedyspeech_params_whole_single_gpu.txt index 9caeb18e..523b5c6e 100644 --- a/tests/chains/speedyspeech/speedyspeech_params_whole_single_gpu.txt +++ b/tests/chains/speedyspeech/speedyspeech_params_whole_single_gpu.txt @@ -13,7 +13,7 @@ null:null null:null ## trainer:norm_train -norm_train:../../../parakeet/exps/speedyspeech/train.py --train-metadata=train_data/processed_BZNSYP/train/norm/metadata.jsonl --dev-metadata=train_data/processed_BZNSYP/dev/norm/metadata.jsonl --config=../../../examples/csmsc/tts2/conf/default.yaml --output-dir=exp/whole --phones-dict=train_data/processed_BZNSYP/phone_id_map.txt --tones-dict=train_data/processed_BZNSYP/tone_id_map.txt --use-relative-path=True +norm_train:../../../paddlespeech/t2s/exps/speedyspeech/train.py --train-metadata=train_data/processed_BZNSYP/train/norm/metadata.jsonl --dev-metadata=train_data/processed_BZNSYP/dev/norm/metadata.jsonl --config=../../../examples/csmsc/tts2/conf/default.yaml --output-dir=exp/whole --phones-dict=train_data/processed_BZNSYP/phone_id_map.txt --tones-dict=train_data/processed_BZNSYP/tone_id_map.txt --use-relative-path=True null:null null:null null:null @@ -21,7 +21,7 @@ null:null null:null ## ===========================eval_params=========================== -eval:../../../parakeet/exps/speedyspeech/synthesize_e2e.py --speedyspeech-config=pretrain_models/speedyspeech_nosil_baker_ckpt_0.5/default.yaml --speedyspeech-checkpoint=pretrain_models/speedyspeech_nosil_baker_ckpt_0.5/snapshot_iter_11400.pdz --speedyspeech-stat=pretrain_models/speedyspeech_nosil_baker_ckpt_0.5/feats_stats.npy --pwg-config=pretrain_models/pwg_baker_ckpt_0.4/pwg_default.yaml --pwg-checkpoint=pretrain_models/pwg_baker_ckpt_0.4/pwg_snapshot_iter_400000.pdz --pwg-stat=pretrain_models/pwg_baker_ckpt_0.4/pwg_stats.npy --text=../../../parakeet/exps/sentences.txt --output-dir=e2e --inference-dir=inference --device="gpu" --phones-dict=pretrain_models/speedyspeech_nosil_baker_ckpt_0.5/phone_id_map.txt --tones-dict=pretrain_models/speedyspeech_nosil_baker_ckpt_0.5/tone_id_map.txt +eval:../../../paddlespeech/t2s/exps/speedyspeech/synthesize_e2e.py --speedyspeech-config=pretrain_models/speedyspeech_nosil_baker_ckpt_0.5/default.yaml --speedyspeech-checkpoint=pretrain_models/speedyspeech_nosil_baker_ckpt_0.5/snapshot_iter_11400.pdz --speedyspeech-stat=pretrain_models/speedyspeech_nosil_baker_ckpt_0.5/feats_stats.npy --pwg-config=pretrain_models/pwg_baker_ckpt_0.4/pwg_default.yaml --pwg-checkpoint=pretrain_models/pwg_baker_ckpt_0.4/pwg_snapshot_iter_400000.pdz --pwg-stat=pretrain_models/pwg_baker_ckpt_0.4/pwg_stats.npy --text=../../../paddlespeech/t2s/exps/sentences.txt --output-dir=e2e --inference-dir=inference --device="gpu" --phones-dict=pretrain_models/speedyspeech_nosil_baker_ckpt_0.5/phone_id_map.txt --tones-dict=pretrain_models/speedyspeech_nosil_baker_ckpt_0.5/tone_id_map.txt null:null ## ===========================infer_params=========================== @@ -37,7 +37,7 @@ null:null null:null null:null null:null -inference:../../../parakeet/exps/speedyspeech/inference.py --inference-dir=pretrain_models/speedyspeech_pwg_inference_0.5 --text=../../../parakeet/exps/sentences.txt --output-dir=inference_out --enable-auto-log --phones-dict=pretrain_models/speedyspeech_pwg_inference_0.5/phone_id_map.txt --tones-dict=pretrain_models/speedyspeech_pwg_inference_0.5/tone_id_map.txt +inference:../../../paddlespeech/t2s/exps/speedyspeech/inference.py --inference-dir=pretrain_models/speedyspeech_pwg_inference_0.5 --text=../../../paddlespeech/t2s/exps/sentences.txt --output-dir=inference_out --enable-auto-log --phones-dict=pretrain_models/speedyspeech_pwg_inference_0.5/phone_id_map.txt --tones-dict=pretrain_models/speedyspeech_pwg_inference_0.5/tone_id_map.txt null:null null:null null:null diff --git a/tests/unit/asr/deepspeech2_model_test.py b/tests/unit/asr/deepspeech2_model_test.py index 00df8195..5835445d 100644 --- a/tests/unit/asr/deepspeech2_model_test.py +++ b/tests/unit/asr/deepspeech2_model_test.py @@ -16,7 +16,7 @@ import unittest import numpy as np import paddle -from deepspeech.models.ds2 import DeepSpeech2Model +from paddlespeech.s2t.models.ds2 import DeepSpeech2Model class TestDeepSpeech2Model(unittest.TestCase): diff --git a/tests/unit/asr/deepspeech2_online_model_test.py b/tests/unit/asr/deepspeech2_online_model_test.py index 6264070b..f623c5ac 100644 --- a/tests/unit/asr/deepspeech2_online_model_test.py +++ b/tests/unit/asr/deepspeech2_online_model_test.py @@ -16,7 +16,7 @@ import unittest import numpy as np import paddle -from deepspeech.models.ds2_online import DeepSpeech2ModelOnline +from paddlespeech.s2t.models.ds2_online import DeepSpeech2ModelOnline class TestDeepSpeech2ModelOnline(unittest.TestCase): diff --git a/tests/unit/asr/error_rate_test.py b/tests/unit/asr/error_rate_test.py index aef25a2c..e5f262c7 100644 --- a/tests/unit/asr/error_rate_test.py +++ b/tests/unit/asr/error_rate_test.py @@ -14,7 +14,7 @@ """Test error rate.""" import unittest -from deepspeech.utils import error_rate +from paddlespeech.s2t.utils import error_rate class TestParse(unittest.TestCase): diff --git a/tests/unit/asr/mask_test.py b/tests/unit/asr/mask_test.py index f44aca8f..d6935997 100644 --- a/tests/unit/asr/mask_test.py +++ b/tests/unit/asr/mask_test.py @@ -16,8 +16,8 @@ import unittest import numpy as np import paddle -from deepspeech.modules.mask import make_non_pad_mask -from deepspeech.modules.mask import make_pad_mask +from paddlespeech.s2t.modules.mask import make_non_pad_mask +from paddlespeech.s2t.modules.mask import make_pad_mask class TestU2Model(unittest.TestCase): diff --git a/tests/unit/asr/u2_model_test.py b/tests/unit/asr/u2_model_test.py index 58600e6b..5b11d2ad 100644 --- a/tests/unit/asr/u2_model_test.py +++ b/tests/unit/asr/u2_model_test.py @@ -17,8 +17,8 @@ import numpy as np import paddle from yacs.config import CfgNode as CN -from deepspeech.models.u2 import U2Model -from deepspeech.utils.layer_tools import summary +from paddlespeech.s2t.models.u2 import U2Model +from paddlespeech.s2t.utils.layer_tools import summary class TestU2Model(unittest.TestCase): diff --git a/tests/unit/tts/test_data_table.py b/tests/unit/tts/test_data_table.py index 3664ea31..3ff5bc1a 100644 --- a/tests/unit/tts/test_data_table.py +++ b/tests/unit/tts/test_data_table.py @@ -11,7 +11,7 @@ # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. -from parakeet.datasets.data_tabel import DataTable +from paddlespeech.t2s.datasets.data_tabel import DataTable def test_audio_dataset(): diff --git a/tests/unit/tts/test_expansion.py b/tests/unit/tts/test_expansion.py index 418e9ba1..db626e3c 100644 --- a/tests/unit/tts/test_expansion.py +++ b/tests/unit/tts/test_expansion.py @@ -13,7 +13,7 @@ # limitations under the License. import paddle -from parakeet.modules import expansion +from paddlespeech.t2s.modules import expansion def test_expand(): diff --git a/tests/unit/tts/test_pwg.py b/tests/unit/tts/test_pwg.py index 2f07a4ab..78cb34f2 100644 --- a/tests/unit/tts/test_pwg.py +++ b/tests/unit/tts/test_pwg.py @@ -18,13 +18,13 @@ from parallel_wavegan.layers import upsample from parallel_wavegan.models import parallel_wavegan as pwgan from timer import timer -from parakeet.models.parallel_wavegan import ConvInUpsampleNet -from parakeet.models.parallel_wavegan import PWGDiscriminator -from parakeet.models.parallel_wavegan import PWGGenerator -from parakeet.models.parallel_wavegan import ResidualBlock -from parakeet.models.parallel_wavegan import ResidualPWGDiscriminator -from parakeet.utils.layer_tools import summary -from parakeet.utils.profile import synchronize +from paddlespeech.t2s.models.parallel_wavegan import ConvInUpsampleNet +from paddlespeech.t2s.models.parallel_wavegan import PWGDiscriminator +from paddlespeech.t2s.models.parallel_wavegan import PWGGenerator +from paddlespeech.t2s.models.parallel_wavegan import ResidualBlock +from paddlespeech.t2s.models.parallel_wavegan import ResidualPWGDiscriminator +from paddlespeech.t2s.utils.layer_tools import summary +from paddlespeech.t2s.utils.profile import synchronize paddle.set_device("gpu:0") device = torch.device("cuda:0") diff --git a/tests/unit/tts/test_reporter.py b/tests/unit/tts/test_reporter.py index bba81d62..a1d83f1f 100644 --- a/tests/unit/tts/test_reporter.py +++ b/tests/unit/tts/test_reporter.py @@ -13,9 +13,9 @@ # limitations under the License. import numpy as np -from parakeet.training.reporter import report -from parakeet.training.reporter import scope -from parakeet.training.reporter import Summary +from paddlespeech.t2s.training.reporter import report +from paddlespeech.t2s.training.reporter import scope +from paddlespeech.t2s.training.reporter import Summary def test_reporter_scope(): diff --git a/tests/unit/tts/test_snapshot.py b/tests/unit/tts/test_snapshot.py index e940a81f..6ceff3e5 100644 --- a/tests/unit/tts/test_snapshot.py +++ b/tests/unit/tts/test_snapshot.py @@ -17,9 +17,9 @@ from itertools import count from paddle import nn from paddle.optimizer import Adam -from parakeet.training.extensions.snapshot import Snapshot -from parakeet.training.trainer import Trainer -from parakeet.training.updater import StandardUpdater +from paddlespeech.t2s.training.extensions.snapshot import Snapshot +from paddlespeech.t2s.training.trainer import Trainer +from paddlespeech.t2s.training.updater import StandardUpdater def test_snapshot(): diff --git a/tests/unit/tts/test_stft.py b/tests/unit/tts/test_stft.py index 8e6ce470..d2d56dca 100644 --- a/tests/unit/tts/test_stft.py +++ b/tests/unit/tts/test_stft.py @@ -18,8 +18,8 @@ import torch from parallel_wavegan.losses import stft_loss as sl from scipy import signal -from parakeet.modules.stft_loss import MultiResolutionSTFTLoss -from parakeet.modules.stft_loss import STFT +from paddlespeech.t2s.modules.stft_loss import MultiResolutionSTFTLoss +from paddlespeech.t2s.modules.stft_loss import STFT def test_stft(): diff --git a/utils/apply-cmvn.py b/utils/apply-cmvn.py index f80053fb..b92e58f4 100755 --- a/utils/apply-cmvn.py +++ b/utils/apply-cmvn.py @@ -6,11 +6,11 @@ from distutils.util import strtobool import kaldiio import numpy -from deepspeech.transform.cmvn import CMVN -from deepspeech.utils.cli_readers import file_reader_helper -from deepspeech.utils.cli_utils import get_commandline_args -from deepspeech.utils.cli_utils import is_scipy_wav_style -from deepspeech.utils.cli_writers import file_writer_helper +from paddlespeech.s2t.transform.cmvn import CMVN +from paddlespeech.s2t.utils.cli_readers import file_reader_helper +from paddlespeech.s2t.utils.cli_utils import get_commandline_args +from paddlespeech.s2t.utils.cli_utils import is_scipy_wav_style +from paddlespeech.s2t.utils.cli_writers import file_writer_helper def get_parser(): diff --git a/utils/build_vocab.py b/utils/build_vocab.py index 67c22fbb..6a903147 100755 --- a/utils/build_vocab.py +++ b/utils/build_vocab.py @@ -21,14 +21,14 @@ import os import tempfile from collections import Counter -from deepspeech.frontend.featurizer.text_featurizer import TextFeaturizer -from deepspeech.frontend.utility import BLANK -from deepspeech.frontend.utility import read_manifest -from deepspeech.frontend.utility import SOS -from deepspeech.frontend.utility import SPACE -from deepspeech.frontend.utility import UNK -from deepspeech.utils.utility import add_arguments -from deepspeech.utils.utility import print_arguments +from paddlespeech.s2t.frontend.featurizer.text_featurizer import TextFeaturizer +from paddlespeech.s2t.frontend.utility import BLANK +from paddlespeech.s2t.frontend.utility import read_manifest +from paddlespeech.s2t.frontend.utility import SOS +from paddlespeech.s2t.frontend.utility import SPACE +from paddlespeech.s2t.frontend.utility import UNK +from paddlespeech.s2t.utils.utility import add_arguments +from paddlespeech.s2t.utils.utility import print_arguments parser = argparse.ArgumentParser(description=__doc__) add_arg = functools.partial(add_arguments, argparser=parser) diff --git a/utils/compute-cmvn-stats.py b/utils/compute-cmvn-stats.py index 706d8cd5..276bcd36 100755 --- a/utils/compute-cmvn-stats.py +++ b/utils/compute-cmvn-stats.py @@ -5,11 +5,11 @@ import logging import kaldiio import numpy as np -from deepspeech.transform.transformation import Transformation -from deepspeech.utils.cli_readers import file_reader_helper -from deepspeech.utils.cli_utils import get_commandline_args -from deepspeech.utils.cli_utils import is_scipy_wav_style -from deepspeech.utils.cli_writers import file_writer_helper +from paddlespeech.s2t.transform.transformation import Transformation +from paddlespeech.s2t.utils.cli_readers import file_reader_helper +from paddlespeech.s2t.utils.cli_utils import get_commandline_args +from paddlespeech.s2t.utils.cli_utils import is_scipy_wav_style +from paddlespeech.s2t.utils.cli_writers import file_writer_helper def get_parser(): diff --git a/utils/compute_mean_std.py b/utils/compute_mean_std.py index 0f63715a..296d272a 100755 --- a/utils/compute_mean_std.py +++ b/utils/compute_mean_std.py @@ -16,11 +16,11 @@ import argparse import functools -from deepspeech.frontend.augmentor.augmentation import AugmentationPipeline -from deepspeech.frontend.featurizer.audio_featurizer import AudioFeaturizer -from deepspeech.frontend.normalizer import FeatureNormalizer -from deepspeech.utils.utility import add_arguments -from deepspeech.utils.utility import print_arguments +from paddlespeech.s2t.frontend.augmentor.augmentation import AugmentationPipeline +from paddlespeech.s2t.frontend.featurizer.audio_featurizer import AudioFeaturizer +from paddlespeech.s2t.frontend.normalizer import FeatureNormalizer +from paddlespeech.s2t.utils.utility import add_arguments +from paddlespeech.s2t.utils.utility import print_arguments parser = argparse.ArgumentParser(description=__doc__) add_arg = functools.partial(add_arguments, argparser=parser) diff --git a/utils/compute_statistics.py b/utils/compute_statistics.py index 84e50ded..e8021c19 100755 --- a/utils/compute_statistics.py +++ b/utils/compute_statistics.py @@ -21,7 +21,7 @@ import numpy as np from sklearn.preprocessing import StandardScaler from tqdm import tqdm -from parakeet.datasets.data_table import DataTable +from paddlespeech.t2s.datasets.data_table import DataTable def main(): diff --git a/utils/copy-feats.py b/utils/copy-feats.py index 7d1b8589..2e120881 100755 --- a/utils/copy-feats.py +++ b/utils/copy-feats.py @@ -3,11 +3,11 @@ import argparse import logging from distutils.util import strtobool -from deepspeech.transform.transformation import Transformation -from deepspeech.utils.cli_readers import file_reader_helper -from deepspeech.utils.cli_utils import get_commandline_args -from deepspeech.utils.cli_utils import is_scipy_wav_style -from deepspeech.utils.cli_writers import file_writer_helper +from paddlespeech.s2t.transform.transformation import Transformation +from paddlespeech.s2t.utils.cli_readers import file_reader_helper +from paddlespeech.s2t.utils.cli_utils import get_commandline_args +from paddlespeech.s2t.utils.cli_utils import is_scipy_wav_style +from paddlespeech.s2t.utils.cli_writers import file_writer_helper def get_parser(): diff --git a/utils/dump_manifest.py b/utils/dump_manifest.py index 4f115590..b5f7b64a 100755 --- a/utils/dump_manifest.py +++ b/utils/dump_manifest.py @@ -17,7 +17,7 @@ import argparse from pathlib import Path from typing import Union -from deepspeech.frontend.utility import read_manifest +from paddlespeech.s2t.frontend.utility import read_manifest key_whitelist = set(['feat', 'text', 'syllable', 'phone']) filename = { diff --git a/utils/feat-to-shape.py b/utils/feat-to-shape.py index 7b36b7e5..bbc9242f 100755 --- a/utils/feat-to-shape.py +++ b/utils/feat-to-shape.py @@ -3,10 +3,10 @@ import argparse import logging import sys -from deepspeech.transform.transformation import Transformation -from deepspeech.utils.cli_readers import file_reader_helper -from deepspeech.utils.cli_utils import get_commandline_args -from deepspeech.utils.cli_utils import is_scipy_wav_style +from paddlespeech.s2t.transform.transformation import Transformation +from paddlespeech.s2t.utils.cli_readers import file_reader_helper +from paddlespeech.s2t.utils.cli_utils import get_commandline_args +from paddlespeech.s2t.utils.cli_utils import is_scipy_wav_style def get_parser(): diff --git a/utils/format_data.py b/utils/format_data.py index 682dbfdb..6fe36997 100755 --- a/utils/format_data.py +++ b/utils/format_data.py @@ -17,11 +17,11 @@ import argparse import functools import json -from deepspeech.frontend.featurizer.text_featurizer import TextFeaturizer -from deepspeech.frontend.utility import load_cmvn -from deepspeech.frontend.utility import read_manifest -from deepspeech.utils.utility import add_arguments -from deepspeech.utils.utility import print_arguments +from paddlespeech.s2t.frontend.featurizer.text_featurizer import TextFeaturizer +from paddlespeech.s2t.frontend.utility import load_cmvn +from paddlespeech.s2t.frontend.utility import read_manifest +from paddlespeech.s2t.utils.utility import add_arguments +from paddlespeech.s2t.utils.utility import print_arguments parser = argparse.ArgumentParser(description=__doc__) add_arg = functools.partial(add_arguments, argparser=parser) diff --git a/utils/format_triplet_data.py b/utils/format_triplet_data.py index f3dd7ca4..79b3d2cb 100755 --- a/utils/format_triplet_data.py +++ b/utils/format_triplet_data.py @@ -17,11 +17,11 @@ import argparse import functools import json -from deepspeech.frontend.featurizer.text_featurizer import TextFeaturizer -from deepspeech.frontend.utility import load_cmvn -from deepspeech.frontend.utility import read_manifest -from deepspeech.utils.utility import add_arguments -from deepspeech.utils.utility import print_arguments +from paddlespeech.s2t.frontend.featurizer.text_featurizer import TextFeaturizer +from paddlespeech.s2t.frontend.utility import load_cmvn +from paddlespeech.s2t.frontend.utility import read_manifest +from paddlespeech.s2t.utils.utility import add_arguments +from paddlespeech.s2t.utils.utility import print_arguments parser = argparse.ArgumentParser(description=__doc__) add_arg = functools.partial(add_arguments, argparser=parser) diff --git a/utils/merge_scp2json.py b/utils/merge_scp2json.py index b724a7dd..650e4669 100755 --- a/utils/merge_scp2json.py +++ b/utils/merge_scp2json.py @@ -8,7 +8,7 @@ import sys from distutils.util import strtobool from io import open -from deepspeech.utils.cli_utils import get_commandline_args +from paddlespeech.s2t.utils.cli_utils import get_commandline_args PY2 = sys.version_info[0] == 2 sys.stdin = codecs.getreader("utf-8")(sys.stdin if PY2 else sys.stdin.buffer) -- GitLab