Commit Graph

  • 162696d4ad Merge pull request #664 from ravi-maithrey/adding_call_for_models_to_contributing Eren Gölge 2021-07-14 15:26:05 +02:00
  • 98d88b88bd added information to ask for model contributions ravi maithrey 2021-07-14 18:16:27 +05:30
  • c79a82ed07 refix linter WeberJulian 2021-07-13 23:12:18 +02:00
  • 7d92b30946 Fix tests WeberJulian 2021-07-13 23:00:34 +02:00
  • 32974dd6a9 Fix test sentences synthesis WeberJulian 2021-07-13 16:04:42 +02:00
  • d906fea08c lint fix and eval as argparse in extract tts spectrograms Edresson 2021-07-13 02:15:31 -03:00
  • 2e5baffa9c Merge fix and eval split as argparse Edresson 2021-07-13 01:47:32 -03:00
  • 93a74cbb71 Merge pull request #628 from Aloento/patch-2 Eren Gölge 2021-07-11 22:17:50 +02:00
  • 639b02fa14 Merge pull request #562 from Sadam1195/patch-1 Eren Gölge 2021-07-11 22:14:08 +02:00
  • d4b1c17b59 Merge pull request #642 from Edresson/main-trainer Eren Gölge 2021-07-11 21:57:19 +02:00
  • 4eac1c4651 bug fix on train_encoder and unit tests Edresson 2021-07-11 12:00:39 -03:00
  • 6e3e6d5756 Change to _get_preprocessor_by_name Aloento 2021-07-08 09:53:13 +02:00
  • 8fbadad68e Bump up to v0.1.2 v0.1.2 Eren Gölge 2021-07-06 14:44:59 +02:00
  • 3c0454490f Fix #616 eren golge 2021-07-06 11:05:05 +02:00
  • 676d22ff0e Merge pull request #611 from coqui-ai/dev v0.1.1 Eren Gölge 2021-07-04 12:59:27 +02:00
  • 0c347624e7 Bump up version to v0.1.1 Eren Gölge 2021-07-04 11:46:36 +02:00
  • 1e9538aaef Add more model tests to test_synthesize Eren Gölge 2021-07-04 11:45:49 +02:00
  • a05b234080 Raise an error when multiple GPUs are in use Eren Gölge 2021-07-04 11:25:49 +02:00
  • 270c3823eb Fix #608 Eren Gölge 2021-07-04 11:19:31 +02:00
  • d00c4c517a Fix #607 Eren Gölge 2021-07-04 11:09:40 +02:00
  • c25a2184e7 Add docs for SpeakerManager v0.1.0 Eren Gölge 2021-07-03 13:55:27 +02:00
  • f382e4c700 Fix linter warnings Eren Gölge 2021-07-03 13:30:24 +02:00
  • 9e7824fe35 Fix UnivNet inference code Eren Gölge 2021-07-02 10:48:34 +02:00
  • 168f97cbe9 Let Synthesizer use the speaker manager out of the model Eren Gölge 2021-07-02 10:47:55 +02:00
  • 196876feb1 Fix ModelManager model download Eren Gölge 2021-07-02 10:47:05 +02:00
  • 9352cb4136 Format Align TTS docstrings Eren Gölge 2021-07-02 10:45:58 +02:00
  • 95ad72f38f Fix glow tts initialization Eren Gölge 2021-07-02 10:45:37 +02:00
  • 40b0b5365e Let get_characters return num_chars Eren Gölge 2021-07-02 10:45:00 +02:00
  • 0fa6a8c9b8 Fix glow tts default parameters Eren Gölge 2021-07-02 10:44:23 +02:00
  • a4c658f5ef Fix for using the Synthesizer out of the model Eren Gölge 2021-07-02 10:43:38 +02:00
  • db47f4f105 Update .models.json Eren Gölge 2021-07-02 10:43:00 +02:00
  • 8cb16da82a Fix readthedocs build Eren Gölge 2021-06-30 15:00:14 +02:00
  • 2e1a428b83 Update glowtts docstrings and docs Eren Gölge 2021-06-30 14:30:55 +02:00
  • 21126839a8 Update .gitignore Eren Gölge 2021-06-30 14:09:49 +02:00
  • 3584a04552 Move main_classes to a separate folder Eren Gölge 2021-06-30 14:09:05 +02:00
  • 5723eb4738 Fix config init in process_args Eren Gölge 2021-06-29 16:41:08 +02:00
  • 4b5421b42f Remove FAQ link from README.md Eren Gölge 2021-06-29 13:20:40 +02:00
  • 47b3b10d6d Bump up to v0.1.0 🚀 Eren Gölge 2021-06-29 13:07:59 +02:00
  • 7ec5c31898 Merge branch 'univnet' into trainer-api Eren Gölge 2021-06-29 10:27:12 +02:00
  • 828e1265a6 Update torch hub Eren Gölge 2021-06-28 16:54:08 +02:00
  • 51398cd15b Add docstrings and typing for audio.py Eren Gölge 2021-06-27 20:56:11 +02:00
  • ae6405bb76 Docstrings for Trainer Eren Gölge 2021-06-28 16:53:57 +02:00
  • 65958eaa41 Add preliminary sphinx documentation Eren Gölge 2021-06-27 20:55:20 +02:00
  • 6b265ae8e3 Docstring update Eren Gölge 2021-06-26 18:34:37 +02:00
  • ab563ce7cd Start training by config.json using register_config Eren Gölge 2021-06-26 18:33:17 +02:00
  • b3c073c99b Allow runing full path scripts with distribute.py Eren Gölge 2021-06-24 19:08:45 +02:00
  • d42d1c02ea Use torch.linalg.qr for pytorch > v1.9.0 Eren Gölge 2021-06-23 13:45:59 +02:00
  • 0a1962b583 Update umap and numba vers. to 0.5.1 and 0.53 Eren Gölge 2021-06-23 11:42:11 +02:00
  • fbba37e01e Fix loading the amp scaler from a checkpoint 🛠️ Eren Gölge 2021-06-23 11:08:34 +02:00
  • a7617d8ab6 Add 🐍 python 3.9 to CI Eren Gölge 2021-06-22 21:39:17 +02:00
  • 9790eddada Fix wrong argument name 🛠️ Eren Gölge 2021-06-22 03:03:30 +02:00
  • 7c614525d9 Add AlignTTS recipe 👩‍🍳 Eren Gölge 2021-06-21 17:23:24 +02:00
  • 932ab107ae Docstring edit in TTSDataset.py ✍️ Eren Gölge 2021-06-21 16:53:19 +02:00
  • cfa5041db7 Fix eval_log for gan.py 🛠️ Eren Gölge 2021-06-21 16:51:28 +02:00
  • d700845b10 Move TorchSTFT to utils.audio Eren Gölge 2021-06-21 16:50:37 +02:00
  • 5b89cb4fec Fixup trainer.py 🛠️ Eren Gölge 2021-06-21 16:49:30 +02:00
  • 418c7d98d5 Create LJSpeech recipes for all the models Eren Gölge 2021-06-18 15:09:40 +02:00
  • 0636c91919 Update gitignore Eren Gölge 2021-06-18 15:02:02 +02:00
  • a196007062 Update Pylint configuration Eren Gölge 2021-06-18 15:01:39 +02:00
  • 8c74f054f0 Enable support for 🐍 python 3.10 Eren Gölge 2021-06-18 14:57:05 +02:00
  • 9455a2b01e Apply small fixes for API compatibility Eren Gölge 2021-06-18 14:54:04 +02:00
  • a5d5bc9063 Print max_decoder_steps when model reaches the limit Eren Gölge 2021-06-18 14:52:27 +02:00
  • e30f245e06 Update synthesizer for speaker and model init Eren Gölge 2021-06-18 14:49:06 +02:00
  • 15fa31b595 fixup configs Eren Gölge 2021-06-18 14:27:28 +02:00
  • f23b228e24 Update speaker_manager Eren Gölge 2021-06-18 14:26:41 +02:00
  • e53616078a Fixup utils for the trainer Eren Gölge 2021-06-18 14:26:11 +02:00
  • 106b63d8a9 Update vocoder utils Eren Gölge 2021-06-18 13:45:52 +02:00
  • 45947acb60 Update TTS.bin scripts for the new API Eren Gölge 2021-06-18 13:44:02 +02:00
  • d7225eedb0 Update vocoder datasets and setup_dataset Eren Gölge 2021-06-18 13:42:36 +02:00
  • d18198dff8 Implement setup_model for vocoder models Eren Gölge 2021-06-18 13:38:01 +02:00
  • e949e7ad58 Update vocoder models Eren Gölge 2021-06-18 13:37:23 +02:00
  • 51005cdab4 Update tts.models.setup_model Eren Gölge 2021-06-18 13:36:47 +02:00
  • 7b8c15ac49 Create base 🐸TTS model abstraction for tts models Eren Gölge 2021-06-18 13:35:36 +02:00
  • a358f74a52 Update vocoder model configs Eren Gölge 2021-06-18 13:29:35 +02:00
  • 786170fe7d Update tts model configs Eren Gölge 2021-06-18 13:28:58 +02:00
  • 626c9d41e6 Update tests for the new trainer API Eren Gölge 2021-06-18 13:27:19 +02:00
  • fcfd95669a Update model test configs Eren Gölge 2021-06-18 13:24:48 +02:00
  • 98298ee671 Implement unified IO utils Eren Gölge 2021-06-18 13:23:08 +02:00
  • c7aad884cd Implement unified trainer Eren Gölge 2021-06-18 13:22:05 +02:00
  • 6d7b5fbcde tts model abstraction with TTSModel Eren Gölge 2021-06-07 19:22:44 +02:00
  • d4dbd89752 fix calculation of loader_start_time Eren Gölge 2021-06-08 17:34:19 +02:00
  • c754a0e17d TrainerAbstract and related updates for TrainerTTS Eren Gölge 2021-06-08 14:18:18 +02:00
  • 00c82c516d rename to Eren Gölge 2021-06-07 16:08:56 +02:00
  • 166f0aeb9a merge if branches with the same implementation Eren Gölge 2021-06-07 15:11:33 +02:00
  • 03494ad642 adjust distribute.py for the train_tts.py Eren Gölge 2021-06-06 15:20:17 +02:00
  • fdfb18d230 downsize melgan test model size Eren Gölge 2021-06-06 13:42:52 +02:00
  • 25238e0658 fix glow-tts inference() Eren Gölge 2021-06-06 13:42:35 +02:00
  • 82582993cc use one testing sentence in tts tests Eren Gölge 2021-06-06 13:41:45 +02:00
  • 419735f440 refactor and fix multi-speaker training in Trainer and Tacotron models Eren Gölge 2021-06-06 13:39:52 +02:00
  • 269e5a734e add max_decoder_steps argument to tacotron models Eren Gölge 2021-06-06 13:38:01 +02:00
  • b3324bd914 fix speaker_manager init Eren Gölge 2021-06-06 13:21:13 +02:00
  • 304d60197b reduce multiband melgan test model size Eren Gölge 2021-06-05 11:48:16 +02:00
  • 2c38ef8441 use get_speaker_manager in Trainer and save speakers.json file when needed Eren Gölge 2021-06-05 11:46:53 +02:00
  • d6b2b6add6 make style and linter fixes Eren Gölge 2021-06-03 13:05:54 +02:00
  • 802d461389 Compute d_vectors and speaker_ids separately in TTSDataset Eren Gölge 2021-06-03 11:46:53 +02:00
  • db6a97d1a2 rename external speaker embedding arguments as d_vectors Eren Gölge 2021-06-03 11:42:40 +02:00
  • 9042ae9195 use to_cuda() for moving data in format_batch() Eren Gölge 2021-06-03 15:05:39 +02:00
  • 877bf66b61 reduce size of the metadata.csv used at testing Eren Gölge 2021-06-02 10:17:58 +02:00
  • f82f1970b8 change to(device) to type_as in models Eren Gölge 2021-06-01 16:26:28 +02:00
  • 9c94b0c5c0 init durations = None Eren Gölge 2021-06-01 14:53:57 +02:00