coqui-ai
/
TTS
mirror of https://github.com/coqui-ai/TTS.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
1,410 Commits
80 Branches
98 Tags
316 MiB
Commit Graph

1410 Commits (a82f7d129d7b5c3068d4c19dc86bf93ce09f330a)

Author SHA1 Message Date
erogol 3b2d726e2d radam pytorch 1.5 update 2020-05-12 13:58:10 +02:00
erogol 574968b249 refactoring utils 2020-05-12 13:57:37 +02:00
erogol 720c4690db update imports 2020-05-12 13:57:26 +02:00
erogol c0c3c6e331 train.py update imports for utils refactoring 2020-05-12 13:57:13 +02:00
erogol 2d9dcd60ba update imports for util refactoring 2020-05-12 13:56:49 +02:00
Eren Gölge 7292d303b9
Merge pull request #404 from Edresson/patch-3 
Fix bug in Graves Attn
 2020-05-07 17:37:25 +02:00
Eren Gölge 7a9247f91f
Merge pull request #403 from Edresson/patch-2 
fix bug in bidirectional decoder train
 2020-05-05 00:33:53 +02:00
Edresson Casanova cce13ee245
Fix bug in Graves Attn 
On my machine at Graves attention the variable self.J ( self.J = torch.arange(0, inputs.shape[1]+2).to(inputs.device) + 0.5) is a LongTensor, but it must be a float tensor. So I get the following error:

Traceback (most recent call last):
  File "train.py", line 704, in <module>
    main(args)
  File "train.py", line 619, in main
    global_step, epoch)
  File "train.py", line 170, in train
    text_input, text_lengths, mel_input, speaker_embeddings=speaker_embeddings)
  File "/home/edresson/anaconda3/envs/TTS2/lib/python3.6/site-packages/torch/nn/modules/module.py", line 489, in __call__
    result = self.forward(*input, **kwargs)
  File "/mnt/edresson/DD/TTS/voice-clonning/TTS/tts_namespace/TTS/models/tacotron.py", line 121, in forward
    self.speaker_embeddings_projected)
  File "/home/edresson/anaconda3/envs/TTS2/lib/python3.6/site-packages/torch/nn/modules/module.py", line 489, in __call__
    result = self.forward(*input, **kwargs)
  File "/mnt/edresson/DD/TTS/voice-clonning/TTS/tts_namespace/TTS/layers/tacotron.py", line 435, in forward
    output, stop_token, attention = self.decode(inputs, mask)
  File "/mnt/edresson/DD/TTS/voice-clonning/TTS/tts_namespace/TTS/layers/tacotron.py", line 367, in decode
    self.attention_rnn_hidden, inputs, self.processed_inputs, mask)
  File "/home/edresson/anaconda3/envs/TTS2/lib/python3.6/site-packages/torch/nn/modules/module.py", line 489, in __call__
    result = self.forward(*input, **kwargs)
  File "/mnt/edresson/DD/TTS/voice-clonning/TTS/tts_namespace/TTS/layers/common_layers.py", line 180, in forward
    phi_t = g_t.unsqueeze(-1) * (1.0 / (1.0 + torch.sigmoid((mu_t.unsqueeze(-1) - j) / sig_t.unsqueeze(-1))))
RuntimeError: expected type torch.cuda.FloatTensor but got torch.cuda.LongTensor


In addition the + 0.5 operation is canceled if it is a LongTensor.
Test: 
>>> torch.arange(0, 10) 
tensor([0, 1, 2, 3, 4, 5, 6, 7, 8, 9])
>>> torch.arange(0, 10) + 0.5
tensor([0, 1, 2, 3, 4, 5, 6, 7, 8, 9])
>>> torch.arange(0, 10.0) + 0.5
tensor([0.5000, 1.5000, 2.5000, 3.5000, 4.5000, 5.5000, 6.5000, 7.5000, 8.5000,
        9.5000])

To resolve this I forced the arrange range to float:
self.J = torch.arange(0, inputs.shape[1]+2.0).to(inputs.device) + 0.5
 2020-05-04 17:52:58 -03:00
Edresson Casanova e4e29f716e
fix bug in bidirectional decoder train 2020-05-04 17:39:35 -03:00
erogol 67420eeb86 more console printing formatting 2020-04-29 11:58:51 +02:00
erogol 091711459d remove redundant avg keeper 2020-04-29 11:58:26 +02:00
Eren Gölge 2e2221f146
Merge pull request #399 from fatihkiralioglu/master 
Tacotron1 + wavernn configuration fix
 2020-04-27 12:00:39 +02:00
fatihkiralioglu cc11be06d7
fixing "No space allowed before..." compile errors 
fixing "No space allowed before..." compile errors
 2020-04-27 09:53:52 +03:00
fatihkiralioglu 70a8210283
Tacotron1 + wavernn configuration fix 
Tacotron1 + wavernn configuration: corrected the input format for wavernn vocoder, converted spectrograms to mels
 2020-04-25 15:18:46 +03:00
Eren Gölge f7b1cad9ee
Merge pull request #398 from PNRxA/dev 
numpy to use CPU when using CUDA
 2020-04-25 12:01:04 +02:00
PNRxA 61a1d59ac5 numpy to use CPU when using CUDA 2020-04-25 16:30:19 +10:00
erogol 373a682c08 logging fix 2020-04-24 18:51:15 +02:00
erogol 95385c8797 adding a dummy normalization file for testint mean-var norm 2020-04-23 16:06:12 +02:00
erogol 53b24625a7 Merge branch 'dev' of https://github.com/mozilla/TTS into dev 2020-04-23 15:48:05 +02:00
erogol f63bce89f6 pylint fix 2020-04-23 15:46:45 +02:00
erogol 3673cc1e30 passing reduced losses to loss dict 2020-04-23 15:46:11 +02:00
erogol 6e2c8c6537 update config.json 2020-04-23 14:37:12 +02:00
erogol d5093bf6fb checkpoint log 2020-04-23 14:24:38 +02:00
erogol 0e7ecca33f fancier and more flexible (self adapting to loss_dict) console logging. Fixing multi-gpu loss reduce 2020-04-23 14:14:09 +02:00
erogol 668a695763 bug fixes and consider the fmin fmax plotting specs 2020-04-09 12:28:52 +02:00
Eren Gölge 99be88c338
Merge pull request #388 from mittimithai/patch-3 
Small fix for "Tacotron" use
 2020-04-02 13:07:17 +02:00
erogol 3293d4e05f bug fix to use tacotron 2020-04-02 13:06:19 +02:00
mittimithai 6501369b0a
Small fix for "Tacotron" use 
"Tacotron" won't work without this fix, since the linear spectrograms end up not getting computed
 2020-04-01 11:57:53 -07:00
erogol 391dab45f0 update ExtractTTSSpecs notebook 2020-03-29 23:07:12 +02:00
erogol a678d684a2 bug fix 2020-03-27 14:17:03 +01:00
erogol d5efe040f7 compute stft paddings to correct wav and spec alignment aespecially for vocoder training 2020-03-26 21:10:37 +01:00
erogol 52c0b4e3e1 bug fix addinf gmissing output for synthesis 2020-03-25 01:56:29 +01:00
erogol 9915d79173 return inputs with synthesis 2020-03-24 01:30:58 +01:00
erogol 745cc4e20a audio.py updates 2020-03-24 01:30:46 +01:00
erogol 20dd509430 change the way how ref_level_db is handled in audio.py 2020-03-17 18:24:05 +01:00
erogol 5223678a4b update audio tests more verbose 2020-03-17 18:22:55 +01:00
erogol b9df54adcd bug fix and check cleaner config field by comparing with the list of 
avail functions
 2020-03-17 15:05:13 +01:00
erogol fa795347a9 turkish cleaner and data preprocessor 2020-03-17 14:47:59 +01:00
erogol fd4e6d0245 Merge branch 'mean-var2' of https://github.com/erogol/TTS_experiments into mean-var2 2020-03-17 13:39:27 +01:00
erogol fd9f469ddc visualization updates wrt mean-var scaling 2020-03-17 13:38:57 +01:00
erogol 77f36b65b8 StandardScaler added 2020-03-17 13:38:57 +01:00
erogol cef9f06887 changes of audio.py for mean-vat scaling 2020-03-17 13:38:57 +01:00
erogol 52b0dc39a6 testing mean-var scalingand updating test config 2020-03-17 13:38:57 +01:00
erogol 40cb4a53a6 update test attention notebooks 2020-03-17 13:38:57 +01:00
erogol 25bcbe2887 config update for mean-var scaling 2020-03-17 13:38:57 +01:00
erogol 568c743632 update compute_statistics.py 2020-03-17 13:38:57 +01:00
erogol 2a903888c1 Merge branch 'dev' of https://github.com/mozilla/TTS into dev 2020-03-17 13:38:38 +01:00
erogol 3bbeb43f57 visualization updates wrt mean-var scaling 2020-03-17 13:28:15 +01:00
erogol d7cf34ca34 StandardScaler added 2020-03-17 13:27:53 +01:00
erogol 92ebec01b1 changes of audio.py for mean-vat scaling 2020-03-17 13:27:25 +01:00