提交 · 4cd60d96b6d737743e152397429ca08f421bc201 · mrywhh / Real-Time-Voice-Cloning

13 3月, 2019 1 次提交
- C
  
  Setup the training configuration for ASR · 4cd60d96
  由 Corentin Jemine 提交于 3月 13, 2019
  
  4cd60d96
07 3月, 2019 3 次提交
- C
  
  Included wave-rnn in the git, with trained models. · 6ac1ff26
  由 Corentin Jemine 提交于 3月 07, 2019
  
  6ac1ff26
- C
  
  Modified tacotron2 preprocessing to cut utterances on silences · bfa74de1
  由 Corentin Jemine 提交于 3月 07, 2019
  
  bfa74de1
- C
  
  Removed WaveNet from tacotron2 · ce881823
  由 Corentin Jemine 提交于 3月 07, 2019
  
  ce881823
04 3月, 2019 1 次提交
- C
  
  Added a method for plotting the embeddings. Included it in the UMAP demo · 5aa216f3
  由 Corentin Jemine 提交于 3月 03, 2019
  
  5aa216f3
26 2月, 2019 2 次提交
- C
  
  Set Tacotron to train with one output per step · 43c84139
  由 Corentin Jemine 提交于 2月 25, 2019
  
  43c84139
- C
  
  Cleaning up. Updated README · 9e88e3b1
  由 Corentin Jemine 提交于 2月 25, 2019
  
  9e88e3b1
22 2月, 2019 2 次提交
- C
  
  Inference demo for Tacotron · b7008db2
  由 Corentin Jemine 提交于 2月 22, 2019
  
  b7008db2
- C
  
  First prototype · 0e8571db
  由 Corentin Jemine 提交于 2月 21, 2019
  
  0e8571db
19 2月, 2019 1 次提交
- C
  
  Adapted tacotron to SV2TTS · 8968ffc1
  由 Corentin Jemine 提交于 2月 18, 2019
  
  8968ffc1
18 2月, 2019 2 次提交
- C
  
  Added more datasets in the demo · a869ffb6
  由 Corentin Jemine 提交于 2月 18, 2019
  
  a869ffb6
- C
  
  Prototype of the demo · 954ce498
  由 Corentin Jemine 提交于 2月 18, 2019
  
  954ce498
17 2月, 2019 1 次提交
- C
  
  Improved the demo · 157e4400
  由 Corentin Jemine 提交于 2月 16, 2019
  
  157e4400
16 2月, 2019 1 次提交
- C
  Reverted relative imports, separated the encoder package and the encoder... · 4116e06d
  由 Corentin Jemine 提交于 2月 15, 2019
```
Reverted relative imports, separated the encoder package and the encoder scripts instead. Improved on the umap demo
```
  4116e06d
15 2月, 2019 1 次提交
- C
  
  Fixed error cases of encoder inference · 60e9d6e4
  由 Corentin Jemine 提交于 2月 14, 2019
  
  60e9d6e4
13 2月, 2019 1 次提交

Renamed SV2TTS to encoder. Finished basic interfacing of the encoder. Changed... · 4d21f6b0

由 Corentin Jemine 提交于 2月 13, 2019

Renamed SV2TTS to encoder. Finished basic interfacing of the encoder. Changed all imports of the encoder to be relative imports in order to use it as a package

4d21f6b0

12 2月, 2019 1 次提交
- C
  
  Implemented inference routines for the encoder, added a backup (5.5% EER) · eb75d8e7
  由 Corentin Jemine 提交于 2月 11, 2019
  
  eb75d8e7
11 2月, 2019 1 次提交
- C
  
  Included VC2 as a dataset for the speaker encoder · f8dc5fce
  由 Corentin Jemine 提交于 2月 11, 2019
  
  f8dc5fce
10 2月, 2019 1 次提交
- C
  
  Adapted tacotron2 hyperparameters to LibriSpeech · 5396a54a
  由 Corentin Jemine 提交于 2月 10, 2019
  
  5396a54a
07 2月, 2019 1 次提交
- C
  
  Included Rayhane-mamah/Tacotron-2 to the git, begun injecting speaker embeddings in Tacotron · 04169a38
  由 Corentin Jemine 提交于 2月 07, 2019
  
  04169a38
01 2月, 2019 2 次提交
- C
  
  Removed backups from the git · 12ba55c0
  由 Corentin Jemine 提交于 2月 01, 2019
  
  12ba55c0
- C
  
  Backup: EER 11% on LS-O & VC1 · 010c4ee6
  由 Corentin Jemine 提交于 2月 01, 2019
  
  010c4ee6
31 1月, 2019 1 次提交
- C
  
  Backing up decent results · dfa3a5c5
  由 Corentin Jemine 提交于 1月 31, 2019
  
  dfa3a5c5
30 1月, 2019 1 次提交
- C
  
  Included VoxCeleb1 · b93c52d3
  由 Corentin Jemine 提交于 1月 30, 2019
  
  b93c52d3
25 1月, 2019 2 次提交
- C
  
  Improved the demo a little · c7992aa6
  由 Corentin Jemine 提交于 1月 25, 2019
  
  c7992aa6
- C
  
  Added a rough demo, put the saved models on the git · 2f555e3c
  由 Corentin Jemine 提交于 1月 25, 2019
  
  2f555e3c
17 1月, 2019 1 次提交
- C
  
  Expanded the training to the entire dataset · f9df0226
  由 Corentin Jemine 提交于 1月 17, 2019
  
  f9df0226
16 1月, 2019 3 次提交
- C
  
  Implemented model saving and loading · 6e1918f8
  由 Corentin Jemine 提交于 1月 16, 2019
  
  6e1918f8
- C
  
  Split data and model parameters, cleaned up EER computation · 827da5ca
  由 Corentin Jemine 提交于 1月 16, 2019
  
  827da5ca
- C
  
  The model is now independant of the number of speakers per batch and of utterances per speaker · 56a2e981
  由 Corentin Jemine 提交于 1月 16, 2019
  
  56a2e981
12 1月, 2019 1 次提交
- C
  
  Replaced accuracy with EER · 4f83bcec
  由 Corentin Jemine 提交于 1月 12, 2019
  
  4f83bcec
04 1月, 2019 1 次提交
- C
  
  Changed the way the dataset and dataloader operate, added LR scheduling · 7e50b3ba
  由 Corentin Jemine 提交于 1月 04, 2019
  
  7e50b3ba
03 1月, 2019 2 次提交
- C
  More speed improvements, doc for the model, fixed improper cycling in... · cc8b4156
  由 Corentin Jemine 提交于 1月 03, 2019
```
More speed improvements, doc for the model, fixed improper cycling in random_partial_utterances, new color scale for UMAP
```
  cc8b4156
- C
  
  Various speed improvements, corrected cosine similarity · f48bfe5c
  由 Corentin Jemine 提交于 1月 03, 2019
  
  f48bfe5c
01 1月, 2019 2 次提交
- C
  
  Added accuracy. Replaced softmax with native implementation · 18588636
  由 Corentin Jemine 提交于 1月 01, 2019
  
  18588636
- C
  
  Implemented visualizations for training · 354ae793
  由 Corentin Jemine 提交于 1月 01, 2019
  
  354ae793
24 12月, 2018 1 次提交
- C
  
  Implemented training with GE2E · 719315ac
  由 Corentin Jemine 提交于 12月 24, 2018
  
  719315ac
22 12月, 2018 1 次提交
- C
  
  Implemented the silence removal algorithm · ef282d06
  由 Corentin Jemine 提交于 12月 22, 2018
  
  ef282d06
17 12月, 2018 1 次提交
- C
  
  Made a UI for debugging purposes · 48d0e8ab
  由 Corentin Jemine 提交于 12月 17, 2018
  
  48d0e8ab
16 12月, 2018 1 次提交
- C
  
  Model definition and training stub · 28737973
  由 Corentin Jemine 提交于 12月 16, 2018
  
  28737973

mrywhh / Real-Time-Voice-Cloning 落后 Fork 源项目 12 个版本

mrywhh / Real-Time-Voice-Cloning
落后 Fork 源项目 12 个版本