Skip to content

D
DeepSpeech

PaddlePaddle / DeepSpeech
大约 2 年 前同步成功

通知 210
Star 8425
Fork 1598
D
DeepSpeech
提交 e2bcaee4 编写于 作者: H Hui Zhang
浏览文件
操作

merge deepspeech, parakeet and text_processing into paddlespeech

上级 1d3b8d81
显示空白变更内容
内联 并排
Showing with 1058 addition and 1058 deletion
.mergify.yml
浏览文件 @ e2bcaee4
......@@ -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"]
......
docs/source/asr/models_introduction.md
浏览文件 @ e2bcaee4
......@@ -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.
......
docs/source/conf.py
浏览文件 @ e2bcaee4
......@@ -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'
......
docs/source/index.rst
浏览文件 @ e2bcaee4
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
......
docs/source/tts/advanced_usage.md
浏览文件 @ e2bcaee4
......@@ -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):
......
docs/source/tts/quick_start.md
浏览文件 @ e2bcaee4
......@@ -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")
......
examples/aishell/s0/path.sh
浏览文件 @ e2bcaee4
......@@ -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/${MODEL}/bin
examples/aishell/s1/path.sh
浏览文件 @ e2bcaee4
......@@ -12,7 +12,7 @@ export LD_LIBRARY_PATH=${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
......
examples/aishell3/tts3/path.sh
浏览文件 @ e2bcaee4
......@@ -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}
examples/aishell3/vc0/README.md
浏览文件 @ e2bcaee4
......@@ -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(`$` and `%`) need to be removed. You shoud preprocess the dataset into the format which MFA needs, the texts have the same name with wavs and have the suffix `.lab`.
We use [lexicon.txt](https://github.com/PaddlePaddle/DeepSpeech/blob/develop/parakeet/exps/voice_cloning/tacotron2_ge2e/lexicon.txt) as the lexicon.
We use [lexicon.txt](https://github.com/PaddlePaddle/DeepSpeech/blob/develop/paddlespeech/t2s/exps/voice_cloning/tacotron2_ge2e/lexicon.txt) as the lexicon.
You can download the alignment results from here [alignment_aishell3.tar.gz](https://paddlespeech.bj.bcebos.com/Parakeet/alignment_aishell3.tar.gz), or train your own MFA model reference to [use_mfa example](https://github.com/PaddlePaddle/DeepSpeech/tree/develop/examples/other/use_mfa) (use MFA1.x now) of our repo.
......
examples/aishell3/vc0/path.sh
浏览文件 @ e2bcaee4
......@@ -10,4 +10,4 @@ export PYTHONIOENCODING=UTF-8
export PYTHONPATH=${MAIN_ROOT}:${PYTHONPATH}
MODEL=voice_cloning/tacotron2_ge2e
export BIN_DIR=${MAIN_ROOT}/parakeet/exps/${MODEL}
export BIN_DIR=${MAIN_ROOT}/paddlespeech/t2s/exps/${MODEL}
examples/callcenter/s1/path.sh
浏览文件 @ e2bcaee4
......@@ -12,4 +12,4 @@ export LD_LIBRARY_PATH=${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
examples/csmsc/tts2/path.sh
浏览文件 @ e2bcaee4
......@@ -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
examples/csmsc/tts3/path.sh
浏览文件 @ e2bcaee4
......@@ -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}
examples/csmsc/voc1/path.sh
浏览文件 @ e2bcaee4
......@@ -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
examples/csmsc/voc3/path.sh
浏览文件 @ e2bcaee4
......@@ -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
examples/librispeech/s0/path.sh
浏览文件 @ e2bcaee4
......@@ -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/${MODEL}/bin
examples/librispeech/s1/path.sh
浏览文件 @ e2bcaee4
......@@ -12,4 +12,4 @@ export LD_LIBRARY_PATH=${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
examples/librispeech/s2/path.sh
浏览文件 @ e2bcaee4
......@@ -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
......
examples/ljspeech/tts0/path.sh
浏览文件 @ e2bcaee4
......@@ -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}
examples/ljspeech/tts1/path.sh
浏览文件 @ e2bcaee4
......@@ -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}
examples/ljspeech/tts3/path.sh
浏览文件 @ e2bcaee4
......@@ -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}
examples/ljspeech/voc0/path.sh
浏览文件 @ e2bcaee4
......@@ -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
examples/ljspeech/voc1/path.sh
浏览文件 @ e2bcaee4
......@@ -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
examples/other/1xt2x/src_deepspeech2x/__init__.py
浏览文件 @ e2bcaee4
......@@ -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()
......
examples/other/1xt2x/src_deepspeech2x/bin/test.py
浏览文件 @ e2bcaee4
......@@ -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):
......
examples/other/1xt2x/src_deepspeech2x/models/ds2/deepspeech2.py
浏览文件 @ e2bcaee4
......@@ -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']
......
examples/other/1xt2x/src_deepspeech2x/models/ds2/rnn.py
浏览文件 @ e2bcaee4
......@@ -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']
......
examples/other/1xt2x/src_deepspeech2x/test_model.py
浏览文件 @ e2bcaee4
......@@ -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()
......
examples/other/ge2e/path.sh
浏览文件 @ e2bcaee4
......@@ -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}
examples/other/text_frontend/test_g2p.py
浏览文件 @ e2bcaee4
......@@ -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"}
......
examples/other/text_frontend/test_textnorm.py
浏览文件 @ e2bcaee4
......@@ -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
......
examples/ted_en_zh/t0/path.sh
浏览文件 @ e2bcaee4
......@@ -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/${MODEL}/bin
examples/timit/s1/path.sh
浏览文件 @ e2bcaee4
......@@ -12,4 +12,4 @@ export LD_LIBRARY_PATH=${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
examples/tiny/s0/path.sh
浏览文件 @ e2bcaee4
......@@ -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/${MODEL}/bin
examples/tiny/s1/path.sh
浏览文件 @ e2bcaee4
......@@ -11,4 +11,4 @@ export LD_LIBRARY_PATH=${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
examples/vctk/tts3/path.sh
浏览文件 @ e2bcaee4
......@@ -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}
examples/vctk/voc1/path.sh
浏览文件 @ e2bcaee4
......@@ -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
hub/setup_hub.sh
浏览文件 @ e2bcaee4
......@@ -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
......
deepspeech/__init__.py → paddlespeech/s2t/__init__.py
浏览文件 @ e2bcaee4
......@@ -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()
......
deepspeech/decoders/beam_search/beam_search.py → paddlespeech/s2t/decoders/beam_search/beam_search.py
浏览文件 @ e2bcaee4
......@@ -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()
......
deepspeech/decoders/ctcdecoder/tests/test_decoders.py → paddlespeech/s2t/decoders/ctcdecoder/tests/test_decoders.py
浏览文件 @ e2bcaee4
......@@ -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):
......
deepspeech/decoders/recog.py → paddlespeech/s2t/decoders/recog.py
浏览文件 @ e2bcaee4
......@@ -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()
......
deepspeech/decoders/recog_bin.py → paddlespeech/s2t/decoders/recog_bin.py
浏览文件 @ e2bcaee4
......@@ -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")
......
deepspeech/decoders/scorers/ctc.py → paddlespeech/s2t/decoders/scorers/ctc.py
浏览文件 @ e2bcaee4
......@@ -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.
"""
......
deepspeech/decoders/utils.py → paddlespeech/s2t/decoders/utils.py
浏览文件 @ e2bcaee4
......@@ -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"]
......
deepspeech/exps/__init__.py → paddlespeech/s2t/exps/__init__.py
浏览文件 @ e2bcaee4
......@@ -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",
}
......
deepspeech/exps/deepspeech2/bin/deploy/client.py → paddlespeech/s2t/exps/deepspeech2/bin/deploy/client.py
浏览文件 @ e2bcaee4
......@@ -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(
......
deepspeech/exps/deepspeech2/bin/deploy/runtime.py → paddlespeech/s2t/exps/deepspeech2/bin/deploy/runtime.py
浏览文件 @ e2bcaee4
......@@ -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):
......
deepspeech/exps/deepspeech2/bin/deploy/send.py → paddlespeech/s2t/exps/deepspeech2/bin/deploy/send.py
浏览文件 @ e2bcaee4
......@@ -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(
......
deepspeech/exps/deepspeech2/bin/deploy/server.py → paddlespeech/s2t/exps/deepspeech2/bin/deploy/server.py
浏览文件 @ e2bcaee4
......@@ -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):
......
deepspeech/exps/deepspeech2/bin/export.py → paddlespeech/s2t/exps/deepspeech2/bin/export.py
浏览文件 @ e2bcaee4
......@@ -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):
......
deepspeech/exps/deepspeech2/bin/test.py → paddlespeech/s2t/exps/deepspeech2/bin/test.py
浏览文件 @ e2bcaee4
......@@ -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):
......
deepspeech/exps/deepspeech2/bin/test_export.py → paddlespeech/s2t/exps/deepspeech2/bin/test_export.py
浏览文件 @ e2bcaee4
......@@ -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):
......
deepspeech/exps/deepspeech2/bin/test_hub.py → paddlespeech/s2t/exps/deepspeech2/bin/test_hub.py
浏览文件 @ e2bcaee4
......@@ -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()
......
deepspeech/exps/deepspeech2/bin/train.py → paddlespeech/s2t/exps/deepspeech2/bin/train.py
浏览文件 @ e2bcaee4
......@@ -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):
......
deepspeech/exps/deepspeech2/config.py → paddlespeech/s2t/exps/deepspeech2/config.py
浏览文件 @ e2bcaee4
......@@ -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'):
......
deepspeech/exps/deepspeech2/model.py → paddlespeech/s2t/exps/deepspeech2/model.py
浏览文件 @ e2bcaee4
......@@ -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()
......
deepspeech/exps/lm/transformer/bin/cacu_perplexity.py → paddlespeech/s2t/exps/lm/transformer/bin/cacu_perplexity.py
浏览文件 @ e2bcaee4
......@@ -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)
......
deepspeech/exps/lm/transformer/lm_cacu_perplexity.py → paddlespeech/s2t/exps/lm/transformer/lm_cacu_perplexity.py
浏览文件 @ e2bcaee4
......@@ -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()
......
deepspeech/exps/u2/bin/alignment.py → paddlespeech/s2t/exps/u2/bin/alignment.py
浏览文件 @ e2bcaee4
......@@ -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):
......
deepspeech/exps/u2_st/bin/export.py → paddlespeech/s2t/exps/u2/bin/export.py
浏览文件 @ e2bcaee4
......@@ -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.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):
......
deepspeech/exps/u2/bin/test.py → paddlespeech/s2t/exps/u2/bin/test.py
浏览文件 @ e2bcaee4
......@@ -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
......
deepspeech/exps/u2/bin/test_hub.py → paddlespeech/s2t/exps/u2/bin/test_hub.py
浏览文件 @ e2bcaee4
......@@ -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
......
deepspeech/exps/u2/bin/train.py → paddlespeech/s2t/exps/u2/bin/train.py
浏览文件 @ e2bcaee4
......@@ -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):
......
deepspeech/exps/u2/config.py → paddlespeech/s2t/exps/u2/config.py
浏览文件 @ e2bcaee4
......@@ -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()
......
deepspeech/exps/u2/model.py → paddlespeech/s2t/exps/u2/model.py
浏览文件 @ e2bcaee4
......@@ -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)
......
deepspeech/exps/u2/trainer.py → paddlespeech/s2t/exps/u2/trainer.py
浏览文件 @ e2bcaee4
......@@ -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()
......
deepspeech/exps/u2_kaldi/bin/recog.py → paddlespeech/s2t/exps/u2_kaldi/bin/recog.py
浏览文件 @ e2bcaee4
......@@ -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:])
deepspeech/exps/u2_kaldi/bin/test.py → paddlespeech/s2t/exps/u2_kaldi/bin/test.py
浏览文件 @ e2bcaee4
......@@ -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",
}
......
deepspeech/exps/u2_kaldi/bin/train.py → paddlespeech/s2t/exps/u2_kaldi/bin/train.py
浏览文件 @ e2bcaee4
......@@ -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",
}
......
deepspeech/exps/u2_kaldi/model.py → paddlespeech/s2t/exps/u2_kaldi/model.py
浏览文件 @ e2bcaee4
......@@ -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)
......
deepspeech/exps/u2/bin/export.py → paddlespeech/s2t/exps/u2_st/bin/export.py
浏览文件 @ e2bcaee4
......@@ -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_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):
......
deepspeech/exps/u2_st/bin/test.py → paddlespeech/s2t/exps/u2_st/bin/test.py
浏览文件 @ e2bcaee4
......@@ -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
......
deepspeech/exps/u2_st/bin/train.py → paddlespeech/s2t/exps/u2_st/bin/train.py
浏览文件 @ e2bcaee4
......@@ -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):
......
deepspeech/exps/u2_st/config.py → paddlespeech/s2t/exps/u2_st/config.py
浏览文件 @ e2bcaee4
......@@ -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()
......
deepspeech/exps/u2_st/model.py → paddlespeech/s2t/exps/u2_st/model.py
浏览文件 @ e2bcaee4
......@@ -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)
......
deepspeech/frontend/augmentor/augmentation.py → paddlespeech/s2t/frontend/augmentor/augmentation.py
浏览文件 @ e2bcaee4
......@@ -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():
......
deepspeech/frontend/augmentor/impulse_response.py → paddlespeech/s2t/frontend/augmentor/impulse_response.py
浏览文件 @ e2bcaee4
......@@ -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):
......
deepspeech/frontend/augmentor/noise_perturb.py → paddlespeech/s2t/frontend/augmentor/noise_perturb.py
浏览文件 @ e2bcaee4
......@@ -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):
......
deepspeech/frontend/augmentor/online_bayesian_normalization.py → paddlespeech/s2t/frontend/augmentor/online_bayesian_normalization.py
浏览文件 @ e2bcaee4
......@@ -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):
......
deepspeech/frontend/augmentor/resample.py → paddlespeech/s2t/frontend/augmentor/resample.py
浏览文件 @ e2bcaee4
......@@ -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):
......
deepspeech/frontend/augmentor/shift_perturb.py → paddlespeech/s2t/frontend/augmentor/shift_perturb.py
浏览文件 @ e2bcaee4
......@@ -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):
......
deepspeech/frontend/augmentor/spec_augment.py → paddlespeech/s2t/frontend/augmentor/spec_augment.py
浏览文件 @ e2bcaee4
......@@ -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()
......
deepspeech/frontend/augmentor/speed_perturb.py → paddlespeech/s2t/frontend/augmentor/speed_perturb.py
浏览文件 @ e2bcaee4
......@@ -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):
......
deepspeech/frontend/augmentor/volume_perturb.py → paddlespeech/s2t/frontend/augmentor/volume_perturb.py
浏览文件 @ e2bcaee4
......@@ -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):
......
deepspeech/frontend/featurizer/speech_featurizer.py → paddlespeech/s2t/frontend/featurizer/speech_featurizer.py
浏览文件 @ e2bcaee4
......@@ -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():
......
deepspeech/frontend/featurizer/text_featurizer.py → paddlespeech/s2t/frontend/featurizer/text_featurizer.py
浏览文件 @ e2bcaee4
......@@ -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()
......
deepspeech/frontend/normalizer.py → paddlespeech/s2t/frontend/normalizer.py
浏览文件 @ e2bcaee4
......@@ -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"]
......
deepspeech/frontend/speech.py → paddlespeech/s2t/frontend/speech.py
浏览文件 @ e2bcaee4
......@@ -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):
......
deepspeech/frontend/utility.py → paddlespeech/s2t/frontend/utility.py
浏览文件 @ e2bcaee4
......@@ -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()
......
deepspeech/io/batchfy.py → paddlespeech/s2t/io/batchfy.py
浏览文件 @ e2bcaee4
......@@ -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"]
......
deepspeech/io/collator.py → paddlespeech/s2t/io/collator.py
浏览文件 @ e2bcaee4
......@@ -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"]
......
deepspeech/io/converter.py → paddlespeech/s2t/io/converter.py
浏览文件 @ e2bcaee4
......@@ -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"]
......
deepspeech/io/dataloader.py → paddlespeech/s2t/io/dataloader.py
浏览文件 @ e2bcaee4
......@@ -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"]
......
deepspeech/io/dataset.py → paddlespeech/s2t/io/dataset.py
浏览文件 @ e2bcaee4
......@@ -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"]
......
deepspeech/io/reader.py → paddlespeech/s2t/io/reader.py
浏览文件 @ e2bcaee4
......@@ -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"]
......
deepspeech/io/sampler.py → paddlespeech/s2t/io/sampler.py
浏览文件 @ e2bcaee4
......@@ -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",
......
deepspeech/io/utility.py → paddlespeech/s2t/io/utility.py
浏览文件 @ e2bcaee4
......@@ -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"]
......
deepspeech/models/asr_interface.py → paddlespeech/s2t/models/asr_interface.py
浏览文件 @ e2bcaee4
......@@ -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",
}
......
deepspeech/models/ds2/conv.py → paddlespeech/s2t/models/ds2/conv.py
浏览文件 @ e2bcaee4
......@@ -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()
......
deepspeech/models/ds2/deepspeech2.py → paddlespeech/s2t/models/ds2/deepspeech2.py
浏览文件 @ e2bcaee4
......@@ -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()
......
deepspeech/models/ds2/rnn.py → paddlespeech/s2t/models/ds2/rnn.py
浏览文件 @ e2bcaee4
......@@ -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()
......
deepspeech/models/ds2_online/conv.py → paddlespeech/s2t/models/ds2_online/conv.py
浏览文件 @ e2bcaee4
......@@ -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):
......
deepspeech/models/ds2_online/deepspeech2.py → paddlespeech/s2t/models/ds2_online/deepspeech2.py
浏览文件 @ e2bcaee4
......@@ -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']
......
deepspeech/models/lm/transformer.py → paddlespeech/s2t/models/lm/transformer.py
浏览文件 @ e2bcaee4
......@@ -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()
......
deepspeech/models/lm_interface.py → paddlespeech/s2t/models/lm_interface.py
浏览文件 @ e2bcaee4
......@@ -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",
}
......
deepspeech/models/st_interface.py → paddlespeech/s2t/models/st_interface.py
浏览文件 @ e2bcaee4
......@@ -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",
}
......
deepspeech/models/u2/u2.py → paddlespeech/s2t/models/u2/u2.py
浏览文件 @ e2bcaee4
......@@ -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"]
......
deepspeech/models/u2/updater.py → paddlespeech/s2t/models/u2/updater.py
浏览文件 @ e2bcaee4
......@@ -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()
......
deepspeech/models/u2_st/u2_st.py → paddlespeech/s2t/models/u2_st/u2_st.py
浏览文件 @ e2bcaee4
......@@ -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"]
......
deepspeech/modules/activation.py → paddlespeech/s2t/modules/activation.py
浏览文件 @ e2bcaee4
......@@ -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()
......
deepspeech/modules/attention.py → paddlespeech/s2t/modules/attention.py
浏览文件 @ e2bcaee4
......@@ -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()
......
deepspeech/modules/cmvn.py → paddlespeech/s2t/modules/cmvn.py
浏览文件 @ e2bcaee4
......@@ -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()
......
deepspeech/modules/conformer_convolution.py → paddlespeech/s2t/modules/conformer_convolution.py
浏览文件 @ e2bcaee4
......@@ -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()
......
deepspeech/modules/crf.py → paddlespeech/s2t/modules/crf.py
浏览文件 @ e2bcaee4
......@@ -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()
......
deepspeech/modules/ctc.py → paddlespeech/s2t/modules/ctc.py
浏览文件 @ e2bcaee4
......@@ -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!")
......
deepspeech/modules/decoder.py → paddlespeech/s2t/modules/decoder.py
浏览文件 @ e2bcaee4
......@@ -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()
......
deepspeech/modules/decoder_layer.py → paddlespeech/s2t/modules/decoder_layer.py
浏览文件 @ e2bcaee4
......@@ -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()
......
deepspeech/modules/embedding.py → paddlespeech/s2t/modules/embedding.py
浏览文件 @ e2bcaee4
......@@ -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()
......
deepspeech/modules/encoder.py → paddlespeech/s2t/modules/encoder.py
浏览文件 @ e2bcaee4
......@@ -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()
......
deepspeech/modules/encoder_layer.py → paddlespeech/s2t/modules/encoder_layer.py
浏览文件 @ e2bcaee4
......@@ -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()
......
deepspeech/modules/loss.py → paddlespeech/s2t/modules/loss.py
浏览文件 @ e2bcaee4
......@@ -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()
......
deepspeech/modules/mask.py → paddlespeech/s2t/modules/mask.py
浏览文件 @ e2bcaee4
......@@ -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()
......
deepspeech/modules/positionwise_feed_forward.py → paddlespeech/s2t/modules/positionwise_feed_forward.py
浏览文件 @ e2bcaee4
......@@ -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()
......
deepspeech/modules/subsampling.py → paddlespeech/s2t/modules/subsampling.py
浏览文件 @ e2bcaee4
......@@ -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()
......
deepspeech/training/cli.py → paddlespeech/s2t/training/cli.py
浏览文件 @ e2bcaee4
......@@ -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
......
deepspeech/training/extensions/evaluator.py → paddlespeech/s2t/training/extensions/evaluator.py
浏览文件 @ e2bcaee4
......@@ -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()
......
deepspeech/training/extensions/snapshot.py → paddlespeech/s2t/training/extensions/snapshot.py
浏览文件 @ e2bcaee4
......@@ -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()
......
deepspeech/training/gradclip.py → paddlespeech/s2t/training/gradclip.py
浏览文件 @ e2bcaee4
......@@ -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"]
......
deepspeech/training/optimizer.py → paddlespeech/s2t/training/optimizer.py
浏览文件 @ e2bcaee4
......@@ -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"]
......
deepspeech/training/scheduler.py → paddlespeech/s2t/training/scheduler.py
浏览文件 @ e2bcaee4
......@@ -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"]
......
deepspeech/training/timer.py → paddlespeech/s2t/training/timer.py
浏览文件 @ e2bcaee4
......@@ -14,7 +14,7 @@
import datetime
import time
from deepspeech.utils.log import Log
from paddlespeech.s2t.utils.log import Log
__all__ = ["Timer"]
......
deepspeech/training/trainer.py → paddlespeech/s2t/training/trainer.py
浏览文件 @ e2bcaee4
......@@ -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"]
......
deepspeech/training/updaters/standard_updater.py → paddlespeech/s2t/training/updaters/standard_updater.py
浏览文件 @ e2bcaee4
......@@ -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"]
......
deepspeech/training/updaters/trainer.py → paddlespeech/s2t/training/updaters/trainer.py
浏览文件 @ e2bcaee4
......@@ -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():
......
deepspeech/training/updaters/updater.py → paddlespeech/s2t/training/updaters/updater.py
浏览文件 @ e2bcaee4
......@@ -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"]
......
deepspeech/transform/functional.py → paddlespeech/s2t/transform/functional.py
浏览文件 @ e2bcaee4
......@@ -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):
......
deepspeech/transform/perturb.py → paddlespeech/s2t/transform/perturb.py
浏览文件 @ e2bcaee4
......@@ -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():
......
deepspeech/transform/spec_augment.py → paddlespeech/s2t/transform/spec_augment.py
浏览文件 @ e2bcaee4
......@@ -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"):
......
deepspeech/transform/transformation.py → paddlespeech/s2t/transform/transformation.py
浏览文件 @ e2bcaee4
......@@ -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():
......
deepspeech/utils/checkpoint.py → paddlespeech/s2t/utils/checkpoint.py
浏览文件 @ e2bcaee4
......@@ -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()
......
deepspeech/utils/cli_readers.py → paddlespeech/s2t/utils/cli_readers.py
浏览文件 @ e2bcaee4
......@@ -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(
......
deepspeech/utils/cli_writers.py → paddlespeech/s2t/utils/cli_writers.py
浏览文件 @ e2bcaee4
......@@ -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(
......
deepspeech/utils/ctc_utils.py → paddlespeech/s2t/utils/ctc_utils.py
浏览文件 @ e2bcaee4
......@@ -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()
......
deepspeech/utils/dynamic_import.py → paddlespeech/s2t/utils/dynamic_import.py
浏览文件 @ e2bcaee4
......@@ -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]
......
deepspeech/utils/profiler.py → paddlespeech/s2t/utils/profiler.py
浏览文件 @ e2bcaee4
......@@ -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()
......
deepspeech/utils/socket_server.py → paddlespeech/s2t/utils/socket_server.py
浏览文件 @ e2bcaee4
......@@ -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"]
......
deepspeech/utils/tensor_utils.py → paddlespeech/s2t/utils/tensor_utils.py
浏览文件 @ e2bcaee4
......@@ -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"]
......
deepspeech/utils/utility.py → paddlespeech/s2t/utils/utility.py
浏览文件 @ e2bcaee4
......@@ -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()
......
parakeet/data/__init__.py → paddlespeech/t2s/data/__init__.py
浏览文件 @ e2bcaee4
......@@ -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 *
parakeet/datasets/am_batch_fn.py → paddlespeech/t2s/datasets/am_batch_fn.py
浏览文件 @ e2bcaee4
......@@ -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):
......
parakeet/exps/fastspeech2/inference.py → paddlespeech/t2s/exps/fastspeech2/inference.py
浏览文件 @ e2bcaee4
......@@ -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():
......
parakeet/exps/fastspeech2/multi_spk_synthesize_e2e.py → paddlespeech/t2s/exps/fastspeech2/multi_spk_synthesize_e2e.py
浏览文件 @ e2bcaee4
......@@ -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):
......
parakeet/exps/fastspeech2/multi_spk_synthesize_e2e_en.py → paddlespeech/t2s/exps/fastspeech2/multi_spk_synthesize_e2e_en.py
浏览文件 @ e2bcaee4
......@@ -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):
......
parakeet/exps/fastspeech2/normalize.py → paddlespeech/t2s/exps/fastspeech2/normalize.py
浏览文件 @ e2bcaee4
......@@ -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():
......
parakeet/exps/fastspeech2/preprocess.py → paddlespeech/t2s/exps/fastspeech2/preprocess.py
浏览文件 @ e2bcaee4
......@@ -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],
......
parakeet/exps/fastspeech2/synthesize.py → paddlespeech/t2s/exps/fastspeech2/synthesize.py
浏览文件 @ e2bcaee4
......@@ -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):
......
parakeet/exps/fastspeech2/synthesize_e2e.py → paddlespeech/t2s/exps/fastspeech2/synthesize_e2e.py
浏览文件 @ e2bcaee4
......@@ -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):
......
parakeet/exps/fastspeech2/synthesize_e2e_en.py → paddlespeech/t2s/exps/fastspeech2/synthesize_e2e_en.py
浏览文件 @ e2bcaee4
......@@ -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):
......
parakeet/exps/fastspeech2/synthesize_e2e_melgan.py → paddlespeech/t2s/exps/fastspeech2/synthesize_e2e_melgan.py
浏览文件 @ e2bcaee4
......@@ -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):
......
parakeet/exps/fastspeech2/train.py → paddlespeech/t2s/exps/fastspeech2/train.py
浏览文件 @ e2bcaee4
......@@ -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):
......
parakeet/exps/gan_vocoder/multi_band_melgan/synthesize.py → paddlespeech/t2s/exps/gan_vocoder/multi_band_melgan/synthesize.py
浏览文件 @ e2bcaee4
......@@ -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():
......
parakeet/exps/gan_vocoder/multi_band_melgan/train.py → paddlespeech/t2s/exps/gan_vocoder/multi_band_melgan/train.py
浏览文件 @ e2bcaee4
......@@ -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):
......
parakeet/exps/gan_vocoder/normalize.py → paddlespeech/t2s/exps/gan_vocoder/normalize.py
浏览文件 @ e2bcaee4
......@@ -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():
......
parakeet/exps/gan_vocoder/parallelwave_gan/synthesize.py → paddlespeech/t2s/exps/gan_vocoder/parallelwave_gan/synthesize.py
浏览文件 @ e2bcaee4
......@@ -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():
......
parakeet/exps/gan_vocoder/parallelwave_gan/synthesize_from_wav.py → paddlespeech/t2s/exps/gan_vocoder/parallelwave_gan/synthesize_from_wav.py
浏览文件 @ e2bcaee4
......@@ -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):
......
parakeet/exps/gan_vocoder/parallelwave_gan/train.py → paddlespeech/t2s/exps/gan_vocoder/parallelwave_gan/train.py
浏览文件 @ e2bcaee4
......@@ -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):
......
parakeet/exps/gan_vocoder/preprocess.py → paddlespeech/t2s/exps/gan_vocoder/preprocess.py
浏览文件 @ e2bcaee4
......@@ -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],
......
parakeet/exps/ge2e/dataset_processors.py → paddlespeech/t2s/exps/ge2e/dataset_processors.py
浏览文件 @ e2bcaee4
......@@ -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):
......
parakeet/exps/ge2e/inference.py → paddlespeech/t2s/exps/ge2e/inference.py
浏览文件 @ e2bcaee4
......@@ -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):
......
parakeet/exps/ge2e/preprocess.py → paddlespeech/t2s/exps/ge2e/preprocess.py
浏览文件 @ e2bcaee4
......@@ -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(
......
parakeet/exps/ge2e/speaker_verification_dataset.py → paddlespeech/t2s/exps/ge2e/speaker_verification_dataset.py
浏览文件 @ e2bcaee4
......@@ -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):
......
parakeet/exps/ge2e/train.py → paddlespeech/t2s/exps/ge2e/train.py
浏览文件 @ e2bcaee4
......@@ -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):
......
parakeet/exps/speedyspeech/inference.py → paddlespeech/t2s/exps/speedyspeech/inference.py
浏览文件 @ e2bcaee4
......@@ -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():
......
parakeet/exps/speedyspeech/normalize.py → paddlespeech/t2s/exps/speedyspeech/normalize.py
浏览文件 @ e2bcaee4
......@@ -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():
......
parakeet/exps/speedyspeech/preprocess.py → paddlespeech/t2s/exps/speedyspeech/preprocess.py
浏览文件 @ e2bcaee4
......@@ -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],
......
parakeet/exps/speedyspeech/synthesize.py → paddlespeech/t2s/exps/speedyspeech/synthesize.py
浏览文件 @ e2bcaee4
......@@ -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):
......
parakeet/exps/speedyspeech/synthesize_e2e.py → paddlespeech/t2s/exps/speedyspeech/synthesize_e2e.py
浏览文件 @ e2bcaee4
......@@ -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):
......
parakeet/exps/speedyspeech/train.py → paddlespeech/t2s/exps/speedyspeech/train.py
浏览文件 @ e2bcaee4
......@@ -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):
......
parakeet/exps/tacotron2/ljspeech.py → paddlespeech/t2s/exps/tacotron2/ljspeech.py
浏览文件 @ e2bcaee4
......@@ -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):
......
parakeet/exps/tacotron2/preprocess.py → paddlespeech/t2s/exps/tacotron2/preprocess.py
浏览文件 @ e2bcaee4
......@@ -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):
......
parakeet/exps/tacotron2/synthesize.ipynb → paddlespeech/t2s/exps/tacotron2/synthesize.ipynb
浏览文件 @ e2bcaee4
......@@ -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"
]
},
{
......
parakeet/exps/tacotron2/synthesize.py → paddlespeech/t2s/exps/tacotron2/synthesize.py
浏览文件 @ e2bcaee4
......@@ -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):
......
parakeet/exps/tacotron2/train.py → paddlespeech/t2s/exps/tacotron2/train.py
浏览文件 @ e2bcaee4
......@@ -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):
......
parakeet/exps/transformer_tts/normalize.py → paddlespeech/t2s/exps/transformer_tts/normalize.py
浏览文件 @ e2bcaee4
......@@ -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():
......
parakeet/exps/transformer_tts/preprocess.py → paddlespeech/t2s/exps/transformer_tts/preprocess.py
浏览文件 @ e2bcaee4
......@@ -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):
......
parakeet/exps/transformer_tts/synthesize.py → paddlespeech/t2s/exps/transformer_tts/synthesize.py
浏览文件 @ e2bcaee4
......@@ -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):
......
parakeet/exps/transformer_tts/synthesize_e2e.py → paddlespeech/t2s/exps/transformer_tts/synthesize_e2e.py
浏览文件 @ e2bcaee4
......@@ -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):
......
parakeet/exps/transformer_tts/train.py → paddlespeech/t2s/exps/transformer_tts/train.py
浏览文件 @ e2bcaee4
......@@ -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):
......
parakeet/exps/voice_cloning/tacotron2_ge2e/aishell3.py → paddlespeech/t2s/exps/voice_cloning/tacotron2_ge2e/aishell3.py
浏览文件 @ e2bcaee4
......@@ -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)
......
parakeet/exps/voice_cloning/tacotron2_ge2e/chinese_g2p.py → paddlespeech/t2s/exps/voice_cloning/tacotron2_ge2e/chinese_g2p.py
浏览文件 @ e2bcaee4
......@@ -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]:
......
parakeet/exps/voice_cloning/tacotron2_ge2e/extract_mel.py → paddlespeech/t2s/exps/voice_cloning/tacotron2_ge2e/extract_mel.py
浏览文件 @ e2bcaee4
......@@ -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,
......
parakeet/exps/voice_cloning/tacotron2_ge2e/train.py → paddlespeech/t2s/exps/voice_cloning/tacotron2_ge2e/train.py
浏览文件 @ e2bcaee4
......@@ -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):
......
parakeet/exps/voice_cloning/tacotron2_ge2e/voice_cloning.py → paddlespeech/t2s/exps/voice_cloning/tacotron2_ge2e/voice_cloning.py
浏览文件 @ e2bcaee4
......@@ -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):
......
parakeet/exps/waveflow/ljspeech.py → paddlespeech/t2s/exps/waveflow/ljspeech.py
浏览文件 @ e2bcaee4
......@@ -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):
......
parakeet/exps/waveflow/preprocess.py → paddlespeech/t2s/exps/waveflow/preprocess.py
浏览文件 @ e2bcaee4
......@@ -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):
......
parakeet/exps/waveflow/synthesize.py → paddlespeech/t2s/exps/waveflow/synthesize.py
浏览文件 @ e2bcaee4
......@@ -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):
......
parakeet/exps/waveflow/train.py → paddlespeech/t2s/exps/waveflow/train.py
浏览文件 @ e2bcaee4
......@@ -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):
......
parakeet/frontend/arpabet.py → paddlespeech/t2s/frontend/arpabet.py
浏览文件 @ e2bcaee4
......@@ -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
......
paddlespeech/t2s/frontend/normalizer/__init__.py 0 → 100644
浏览文件 @ e2bcaee4
# 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 *
parakeet/frontend/normalizer/normalizer.py → paddlespeech/t2s/frontend/normalizer/normalizer.py
浏览文件 @ e2bcaee4
......@@ -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):
......
parakeet/frontend/phonectic.py → paddlespeech/t2s/frontend/phonectic.py
浏览文件 @ e2bcaee4
......@@ -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
......
parakeet/frontend/pinyin.py → paddlespeech/t2s/frontend/pinyin.py
浏览文件 @ e2bcaee4
......@@ -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 = [
......
parakeet/frontend/zh_frontend.py → paddlespeech/t2s/frontend/zh_frontend.py
浏览文件 @ e2bcaee4
......@@ -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():
......
paddlespeech/t2s/frontend/zh_normalization/__init__.py 0 → 100644
浏览文件 @ e2bcaee4
# 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 *
parakeet/models/fastspeech2/fastspeech2.py → paddlespeech/t2s/models/fastspeech2/fastspeech2.py
浏览文件 @ e2bcaee4
......@@ -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):
......
parakeet/models/fastspeech2/fastspeech2_updater.py → paddlespeech/t2s/models/fastspeech2/fastspeech2_updater.py
浏览文件 @ e2bcaee4
......@@ -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]')
......
parakeet/models/melgan/melgan.py → paddlespeech/t2s/models/melgan/melgan.py
浏览文件 @ e2bcaee4
......@@ -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):
......
parakeet/models/melgan/multi_band_melgan_updater.py → paddlespeech/t2s/models/melgan/multi_band_melgan_updater.py
浏览文件 @ e2bcaee4
......@@ -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]')
......
parakeet/models/parallel_wavegan/parallel_wavegan_updater.py → paddlespeech/t2s/models/parallel_wavegan/parallel_wavegan_updater.py
浏览文件 @ e2bcaee4
......@@ -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]')
......
parakeet/models/speedyspeech/speedyspeech.py → paddlespeech/t2s/models/speedyspeech/speedyspeech.py
浏览文件 @ e2bcaee4
......@@ -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):
......
parakeet/models/speedyspeech/speedyspeech_updater.py → paddlespeech/t2s/models/speedyspeech/speedyspeech_updater.py
浏览文件 @ e2bcaee4
......@@ -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]')
......
parakeet/models/tacotron2.py → paddlespeech/t2s/models/tacotron2.py
浏览文件 @ e2bcaee4
......@@ -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"]
......
parakeet/models/transformer_tts/transformer_tts.py → paddlespeech/t2s/models/transformer_tts/transformer_tts.py
浏览文件 @ e2bcaee4
......@@ -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):
......
parakeet/models/transformer_tts/transformer_tts_updater.py → paddlespeech/t2s/models/transformer_tts/transformer_tts_updater.py
浏览文件 @ e2bcaee4
......@@ -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]')
......
parakeet/models/waveflow.py → paddlespeech/t2s/models/waveflow.py
浏览文件 @ e2bcaee4
......@@ -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"]
......
parakeet/modules/fastspeech2_predictor/duration_predictor.py → paddlespeech/t2s/modules/fastspeech2_predictor/duration_predictor.py
浏览文件 @ e2bcaee4
......@@ -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):
......
parakeet/modules/fastspeech2_predictor/variance_predictor.py → paddlespeech/t2s/modules/fastspeech2_predictor/variance_predictor.py
浏览文件 @ e2bcaee4
......@@ -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):
......
parakeet/modules/fastspeech2_transformer/attention.py → paddlespeech/t2s/modules/fastspeech2_transformer/attention.py
浏览文件 @ e2bcaee4
......@@ -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):
......
parakeet/modules/fastspeech2_transformer/decoder.py → paddlespeech/t2s/modules/fastspeech2_transformer/decoder.py
浏览文件 @ e2bcaee4
......@@ -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):
......
parakeet/modules/fastspeech2_transformer/decoder_layer.py → paddlespeech/t2s/modules/fastspeech2_transformer/decoder_layer.py
浏览文件 @ e2bcaee4
......@@ -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):
......
parakeet/modules/fastspeech2_transformer/encoder.py → paddlespeech/t2s/modules/fastspeech2_transformer/encoder.py
浏览文件 @ e2bcaee4
......@@ -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):
......
parakeet/modules/fastspeech2_transformer/lightconv.py → paddlespeech/t2s/modules/fastspeech2_transformer/lightconv.py
浏览文件 @ e2bcaee4
......@@ -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)
......
parakeet/modules/residual_stack.py → paddlespeech/t2s/modules/residual_stack.py
浏览文件 @ e2bcaee4
......@@ -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):
......
parakeet/modules/style_encoder.py → paddlespeech/t2s/modules/style_encoder.py
浏览文件 @ e2bcaee4
......@@ -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):
......
parakeet/modules/transformer.py → paddlespeech/t2s/modules/transformer.py
浏览文件 @ e2bcaee4
......@@ -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",
......
parakeet/training/cli.py → paddlespeech/t2s/training/cli.py
浏览文件 @ e2bcaee4
......@@ -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
-------
......
parakeet/training/experiment.py → paddlespeech/t2s/training/experiment.py
浏览文件 @ e2bcaee4
......@@ -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"]
......
parakeet/training/extensions/evaluator.py → paddlespeech/t2s/training/extensions/evaluator.py
浏览文件 @ e2bcaee4
......@@ -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):
......
parakeet/training/extensions/snapshot.py → paddlespeech/t2s/training/extensions/snapshot.py
浏览文件 @ e2bcaee4
......@@ -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):
......
parakeet/training/extensions/visualizer.py → paddlespeech/t2s/training/extensions/visualizer.py
浏览文件 @ e2bcaee4
......@@ -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):
......
parakeet/training/trainer.py → paddlespeech/t2s/training/trainer.py
浏览文件 @ e2bcaee4
......@@ -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):
......
parakeet/training/trigger.py → paddlespeech/t2s/training/trigger.py
浏览文件 @ e2bcaee4
......@@ -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):
......
parakeet/training/updaters/standard_updater.py → paddlespeech/t2s/training/updaters/standard_updater.py
浏览文件 @ e2bcaee4
......@@ -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):
......
parakeet/utils/checkpoint.py → paddlespeech/t2s/utils/checkpoint.py
浏览文件 @ e2bcaee4
......@@ -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"]
......
text_processing/speechtask/punctuation_restoration/utils/default_parser.py → paddlespeech/text/speechtask/punctuation_restoration/utils/default_parser.py
浏览文件 @ e2bcaee4
......@@ -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
......
parakeet/frontend/normalizer/__init__.py 已删除 100644 → 0
浏览文件 @ 1d3b8d81
# 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 *
parakeet/frontend/zh_normalization/__init__.py 已删除 100644 → 0
浏览文件 @ 1d3b8d81
# 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 *
setup.py
浏览文件 @ e2bcaee4
......@@ -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.")
......
setup.sh
浏览文件 @ e2bcaee4
......@@ -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
......
tests/benchmark/pwgan/README.md
浏览文件 @ e2bcaee4
......@@ -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
tests/benchmark/pwgan/run_all.sh
浏览文件 @ e2bcaee4
......@@ -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
......
tests/benchmark/pwgan/run_benchmark.sh
浏览文件 @ e2bcaee4
......@@ -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
......
tests/chains/ds2/ds2_params_lite_train_infer.txt
浏览文件 @ e2bcaee4
......@@ -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
......
tests/chains/ds2/ds2_params_whole_train_infer.txt
浏览文件 @ e2bcaee4
......@@ -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
......
tests/chains/speedyspeech/speedyspeech_params_lite_multi_gpu.txt
浏览文件 @ e2bcaee4
......@@ -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
......
tests/chains/speedyspeech/speedyspeech_params_lite_single_gpu.txt
浏览文件 @ e2bcaee4
......@@ -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
......
tests/chains/speedyspeech/speedyspeech_params_whole_multi_gpu.txt
浏览文件 @ e2bcaee4
......@@ -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
......
tests/chains/speedyspeech/speedyspeech_params_whole_single_gpu.txt
浏览文件 @ e2bcaee4
......@@ -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
......
tests/unit/asr/deepspeech2_model_test.py
浏览文件 @ e2bcaee4
......@@ -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):
......
tests/unit/asr/deepspeech2_online_model_test.py
浏览文件 @ e2bcaee4
......@@ -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):
......
tests/unit/asr/error_rate_test.py
浏览文件 @ e2bcaee4
......@@ -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):
......
tests/unit/asr/mask_test.py
浏览文件 @ e2bcaee4
......@@ -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):
......
tests/unit/asr/u2_model_test.py
浏览文件 @ e2bcaee4
......@@ -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):
......
tests/unit/tts/test_data_table.py
浏览文件 @ e2bcaee4
......@@ -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():
......
tests/unit/tts/test_expansion.py
浏览文件 @ e2bcaee4
......@@ -13,7 +13,7 @@
# limitations under the License.
import paddle
from parakeet.modules import expansion
from paddlespeech.t2s.modules import expansion
def test_expand():
......
tests/unit/tts/test_pwg.py
浏览文件 @ e2bcaee4
......@@ -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")
......
tests/unit/tts/test_reporter.py
浏览文件 @ e2bcaee4
......@@ -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():
......
tests/unit/tts/test_snapshot.py
浏览文件 @ e2bcaee4
......@@ -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():
......
tests/unit/tts/test_stft.py
浏览文件 @ e2bcaee4
......@@ -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():
......
utils/apply-cmvn.py
浏览文件 @ e2bcaee4
......@@ -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():
......
utils/build_vocab.py
浏览文件 @ e2bcaee4
......@@ -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)
......
utils/compute-cmvn-stats.py
浏览文件 @ e2bcaee4
......@@ -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():
......
utils/compute_mean_std.py
浏览文件 @ e2bcaee4
......@@ -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)
......
utils/compute_statistics.py
浏览文件 @ e2bcaee4
......@@ -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():
......
utils/copy-feats.py
浏览文件 @ e2bcaee4
......@@ -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():
......
utils/dump_manifest.py
浏览文件 @ e2bcaee4
......@@ -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 = {
......
utils/feat-to-shape.py
浏览文件 @ e2bcaee4
......@@ -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():
......
utils/format_data.py
浏览文件 @ e2bcaee4
......@@ -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)
......
utils/format_triplet_data.py
浏览文件 @ e2bcaee4
......@@ -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)
......
utils/merge_scp2json.py
浏览文件 @ e2bcaee4
......@@ -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)
......
Markdown is supported
0% 或 .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑！
想要评论请 注册 或