Commit Graph

  • 4001322e50 Fix trailing space WeberJulian 2021-11-27 23:41:55 +01:00
  • 352b4be104 Move multilingual logic out of the trainer WeberJulian 2021-11-27 22:55:21 +01:00
  • b909a3b63e Add Docstring for TorchSTFT Edresson 2021-11-26 17:42:26 -03:00
  • d79531c177 Add remove silence vad script Unit test Edresson 2021-11-26 17:14:58 -03:00
  • c382a3f6a9 Add find unique phonemes unit tests Edresson 2021-11-26 17:14:29 -03:00
  • be8f444636 Add the SCL resample TODO Edresson 2021-11-23 13:37:14 -03:00
  • 4a8c344fc8 Make a multilingual test use chars WeberJulian 2021-11-23 16:29:13 +01:00
  • eff0a5ca10 Fix merge bug WeberJulian 2021-11-23 16:25:21 +01:00
  • 9c1bec86a4 Fix tests WeberJulian 2021-11-23 16:00:38 +01:00
  • 79f75924de Fix pylint checks Edresson 2021-11-23 11:43:08 -03:00
  • 93dbb67c52 Remove self.audio_config from VITS WeberJulian 2021-11-23 15:40:03 +01:00
  • 0359cab4fa Remove torchaudio requeriment Edresson 2021-11-23 11:24:36 -03:00
  • 36ddf32972 Fix trailing whitespace WeberJulian 2021-11-23 15:24:03 +01:00
  • 9d8d4e6fb3 Update docstring WeberJulian 2021-11-23 14:50:19 +01:00
  • 65b5ed4427 Add VITS multilingual d-vectors unit test Edresson 2021-11-22 20:19:37 -03:00
  • 87059e3bbb Add the language embedding dim in the duration predictor class Edresson 2021-11-22 20:02:05 -03:00
  • 5782df8ffe Get the number speaker from the Speaker Manager property Edresson 2021-11-22 18:52:59 -03:00
  • 88f4369c63 Fix the bug in multispeaker vits Edresson 2021-11-22 18:43:56 -03:00
  • aa1a070d58 Rename ununsed_speakers to ignored_speakers Edresson 2021-11-22 18:42:54 -03:00
  • 346bd931b2 remove inference notebook WeberJulian 2021-11-22 22:42:12 +01:00
  • 1251d04387 Fix function name Edresson 2021-11-22 16:41:46 -03:00
  • 9781e4d516 Lint fixs Edresson 2021-11-22 15:55:00 -03:00
  • 8ae22725e1 Fix the bug in extract tts spectrograms Edresson 2021-11-22 13:43:44 -03:00
  • d0cb700e7d Set the new_fields in copy_model_files as None by default Edresson 2021-11-22 09:25:32 -03:00
  • 5ba416e2e8 Remove the call to get_speaker_manager Edresson 2021-11-22 09:18:29 -03:00
  • cad82a9296 Remove the data from the set_d_vectors_from_file function Edresson 2021-11-22 09:05:28 -03:00
  • ec31dacbb7 Remove unusable speaker manager function Edresson 2021-11-22 08:57:13 -03:00
  • 86b2536491 Turn more clear the VITS loss function Edresson 2021-11-22 08:48:56 -03:00
  • 5fc127bb7a Remove the unusable fine-tuning model Edresson 2021-11-22 08:19:36 -03:00
  • 89019d49a2 Create a module for the VAD script Edresson 2021-11-21 12:20:35 -03:00
  • 390096fe0f fix d-vector WeberJulian 2021-11-18 00:17:42 +01:00
  • 868cf6424f Fix small issues WeberJulian 2021-11-15 22:09:59 +01:00
  • e04577575e Fix use_speaker_embedding logic WeberJulian 2021-11-09 12:20:43 +01:00
  • 5f40e96010 Fix continue path WeberJulian 2021-11-09 12:20:11 +01:00
  • 61251bd86c Fix phonemes WeberJulian 2021-11-06 00:27:58 +01:00
  • b1df118b81 fix imports for load_meta_data WeberJulian 2021-11-04 16:36:40 +01:00
  • 686c7381e2 fix phonemes per language WeberJulian 2021-11-04 16:36:11 +01:00
  • 215a74b32e fix linter WeberJulian 2021-11-02 19:10:18 +01:00
  • 3e9ca4b95d make style WeberJulian 2021-11-02 17:31:14 +01:00
  • e1bdeacd2e Add torchaudio in requirements.txt WeberJulian 2021-11-02 17:30:20 +01:00
  • 88d6399e12 fix test sentence synthesis WeberJulian 2021-11-02 16:33:40 +01:00
  • 20ac31dc71 fix f0_cache_path in dataset WeberJulian 2021-10-29 19:05:26 +02:00
  • 6ed55ba57e fix test vits WeberJulian 2021-10-29 17:09:10 +02:00
  • 21b49c3acd fix collate_fn WeberJulian 2021-10-27 13:45:49 +02:00
  • ec83ffbd7a PitchExtractor Julian WEBER 2021-10-27 13:40:11 +02:00
  • 3440c54bbe get_aux_input Julian WEBER 2021-10-27 12:02:02 +02:00
  • 5c89803968 Merge dataset Julian WEBER 2021-10-27 11:54:05 +02:00
  • c80cf67d3d Add remove silence VAD script Edresson 2021-10-26 11:35:18 -03:00
  • 3ac428340d Add audio resample in the speaker consistency loss Edresson 2021-10-19 08:07:48 -03:00
  • 39aff6685e Add freeze vocoder generator and flow-based decoder option Edresson 2021-09-19 21:06:58 -03:00
  • de41165af4 freeze vits parts WeberJulian 2021-09-19 23:35:31 +02:00
  • 9d2c445e3d get_speaker_weighted_sampler WeberJulian 2021-09-19 23:34:38 +02:00
  • 56480360cf Update the VITS model docs Edresson 2021-09-19 13:29:09 -03:00
  • cd7639ca70 Add voice conversion fine tuning mode Edresson 2021-09-14 17:27:00 -03:00
  • 2be38aad3f Added a notbook for d-vector multilingual VITS WeberJulian 2021-07-21 16:49:12 +02:00
  • 3cd889a9d4 Add support to use the speaker encoder as loss function in VITS model Edresson 2021-09-03 07:37:43 -03:00
  • a3901032f4 Add H/ASP original checkpoint support Edresson 2021-09-01 09:23:45 -03:00
  • fee01daa09 Add the ValueError in the restore checkpoint exception to avoid problems with the optimizer restauration when new keys are addition Edresson 2021-08-28 10:11:33 -03:00
  • ecf327a118 Add VITS multispeaker train unit test Edresson 2021-08-28 09:59:48 -03:00
  • 2bba769e67 Active the multispeaker mode in multilingual training Edresson 2021-08-28 09:59:09 -03:00
  • 256197b6aa Fix the optimizer parameters bug in multilingual and multispeaker training Edresson 2021-08-28 09:57:52 -03:00
  • f4abb19515 Fix bug after merge Edresson 2021-08-26 16:01:07 -03:00
  • d7042ecfd8 Fix d-vector multispeaker training bug Edresson 2021-08-26 14:47:53 -03:00
  • 08da902af3 Add VITS d-vector unit test Edresson 2021-08-26 12:19:01 -03:00
  • 859cf1bfac Add VITS multilingual unit test Edresson 2021-08-26 12:18:05 -03:00
  • 82611cfcd3 Fix unit tests Edresson 2021-08-26 11:36:11 -03:00
  • cfa9910f9d Fix pylint issues Edresson 2021-08-26 07:55:41 -03:00
  • 9071bf326f Implement vocoder Fine Tuning like SC-GlowTTS paper Edresson 2021-08-25 16:52:02 -03:00
  • 3df5d9a619 Fix the bug in M-AILABS formatter Edresson 2021-08-23 16:12:31 -03:00
  • d653227e59 Add voice conversion support for the model VITS trained with external speaker embedding Edresson 2021-08-19 14:05:42 -03:00
  • 56b548835d Fix bug in VITS multilingual inference Edresson 2021-08-19 09:59:41 -03:00
  • 240356cd53 Fix bugs in the non-multilingual VITS inference Edresson 2021-08-17 07:41:21 -03:00
  • 32ece5d5ad Fix pylint issues Edresson 2021-08-15 13:44:58 -03:00
  • 8e83a212fa Add multilingual inference support Edresson 2021-08-14 17:52:00 -03:00
  • d0e3647db6 Add multilingual training support to the VITS model Edresson 2021-08-13 21:40:34 -03:00
  • 829ee55b04 Implement multilingual dataloader support Edresson 2021-08-13 19:58:56 -03:00
  • c9f5838bb4 Fix pylint issues Edresson 2021-08-13 03:38:54 -03:00
  • 1efcccd5c9 Implement training support with d_vecs in the VITS model Edresson 2021-08-13 00:49:38 -03:00
  • c9c1960040 Allow ignore speakers for all multispeaker datasets Edresson 2021-08-12 22:48:30 -03:00
  • 234a4aacb3 Select randomly a speaker from the speaker manager for the test setences Edresson 2021-08-12 21:52:12 -03:00
  • 8310d19da8 Save speakers embeddings/ids before starting training Edresson 2021-08-12 07:52:44 -03:00
  • 7f1a23787e Merge pull request #914 from coqui-ai/dev v0.4.2 Eren Gölge 2021-12-08 16:41:44 +01:00
  • bce143c738 server: fix compatibility with tts_models/en/ljspeech/fast_pitch (#893) Jörg Thalheim 2021-12-07 13:36:29 +00:00
  • babdd84f91 Fix GST inference Eren Gölge 2021-12-07 13:28:49 +00:00
  • ce45d9e1af Make style and lint Eren Gölge 2021-12-01 10:42:52 +00:00
  • 40cb8ac966 Fix #958 Eren Gölge 2021-12-01 10:33:34 +00:00
  • 512ada7548 Fix callbacks against multi-gpu training Eren Gölge 2021-12-01 10:32:14 +00:00
  • 9a145c9b88 Documentation corrections for finetuning and data preparation (#931) Baybars Külebi 2021-11-15 18:14:55 +01:00
  • 2ed9e3c241 Fix constant use of noise augment Eren Gölge 2021-11-08 09:20:34 +01:00
  • b6b14a76af Fix VITS stochastic duration predictor Eren Gölge 2021-11-08 09:20:11 +01:00
  • 3a77899775 Update issue and feature request templates Eren Gölge 2021-11-08 09:19:37 +01:00
  • dc3dd55dd9 Add collect_env_info.py Eren Gölge 2021-11-08 08:59:08 +01:00
  • faafea4cf2 Fix style Eren Gölge 2021-11-04 17:04:40 +01:00
  • d227aaebcc Print when using Griffin-Lim in Synthesizer Eren Gölge 2021-11-01 16:52:26 +01:00
  • c5077c6c3f Merge branch 'dev' of https://github.com/coqui-ai/TTS into dev Eren Gölge 2021-11-01 16:42:27 +01:00
  • 20cebde1c9 Add docstring to MAI labs formatter Eren Gölge 2021-11-01 16:41:55 +01:00
  • 608f437545 Add a function to find unique chars Eren Gölge 2021-11-01 16:41:33 +01:00
  • d6d780e758 Fix FastSpeech config Eren Gölge 2021-11-01 16:41:15 +01:00
  • 5ba47081ee Use GL for VCTK FastPitch models Eren Gölge 2021-11-01 16:39:03 +01:00
  • 3bc043faeb Upgrade to gruut 2.0 (#882) Michael Hansen 2021-10-31 06:41:55 -04:00