Releases: idiap/coqui-ai-TTS
Releases · idiap/coqui-ai-TTS
v0.25.3
v0.25.2
What's Changed
Features
- Add kNN-VC model by @eginhard in #256
- Support all Coqui TTS models in the server by @eginhard in #252
- Allow both Path and strings where possible and add type hints by @eginhard in #210
- feat(manager): print download location when listing models by @eginhard in #213
Fixes
- fix(bark): handle broken paths in config by @eginhard in #253
- fix(openvoice): correctly set utterance length by @eginhard in #260
- fix(bin): log to stdout in cli tools by @eginhard in #217
- fix(vc): support both cpu and cuda by @eginhard in #244
- fix(xtts): voice_dir should remain None if not specified by @eginhard in #224
- Fix num2words call using non-standard lang code by @SkaceKamen in #237
- chore: remove unused callback code by @eginhard in #229
- fix: convert >35 digit English numbers digit-by-digit by @lostways in #240
- Change old docker image url to the one that is relevant to this repo in README.md by @DelovoiDC in #243
- test: switch from nose2 to pytest by @eginhard in #208
- Update plot_embeddings_umap notebook by @eginhard in #221
- Improve documentation by @eginhard in #207
New Contributors
- @SkaceKamen made their first contribution in #237
- @lostways made their first contribution in #240
- @DelovoiDC made their first contribution in #243
Full Changelog: v0.25.1...v0.25.2
WavLM-HiFiGAN vocoders from kNN-VC
- HiFiGAN vocoders for WavLM features trained on LibriSpeech100 from https://github.com/bshall/knn-vc (MIT license)
v0.25.1
v0.25.0
What's Changed
Features
Fixes
- Automatically convert audio to mono, add more helpful error messages by @eginhard in #166
- fix(bin.synthesize): return speakers names only by @shavit in #147
- Show original model URLs by @eginhard in #149
- Support for building Docker on arm64 by @hongkongkiwi in #159
- refactor: handle deprecation of torch.cuda.amp.autocast by @eginhard in #144
Dev
- build: move doc dependencies from extra into group and build with uv by @eginhard in #133
- Use external package for monotonic alignment search by @eginhard in #135
- ci: allow testing out trainer/coqpit branches before release by @eginhard in #168
- Remove unused code by @eginhard in #172
- build: switch to forked coqpit by @eginhard in #110
New Contributors
- @hongkongkiwi made their first contribution in #159
- @ajk1402 made their first contribution in #183
Full Changelog: v0.24.3...v0.25.0
v0.24.3
What's Changed
Fixes
- Load weights only in torch.load for pytorch>=2.4 by @shavit in #77 and @eginhard in #113
- Add compatibility with transformers>=4.43 by @JohnnyStreet in #109
- fix(gpt): set attention mask and address other warnings by @eginhard in #114
- fix(text.characters): add nasal diacritic by @eginhard in #127
New Contributors
- @JohnnyStreet made their first contribution in #109
Full Changelog: v0.24.2...v0.24.3
Contributors
shavit, eginhard, and JohnnyStreet
Assets 2
v0.24.2
What's Changed
Highlights
- Prebuilt wheels now also provided for Mac and Windows
- Hindi fully supported for XTTS fine-tuning
Features
- Add multilingual phoneme cleaner by @eginhard in #44
- Support hindi in XTTS tokenizer by @akshatrocky in #64
- Normalize unicode characters in text cleaners by @shavit in #85
Fixes
- Improve CLI handling of boolean arguments by @eginhard in #38
- Use the specified device for pretrained speaker encoder by @ChristianRomberg in #45
- XTTS streaming fixes by @eginhard in #46 and by @gravityrail in #59
- Numpy 2 support by @eginhard in #47 #56 #81
- VC-related refactors and fixes by @eginhard in #49
- Update to coqui-tts-trainer 0.1.4 by @eginhard in #51
- Fix typo in xtts.py by @abrahammathews2000 in #53
- build: move umap-learn into optional notebook dependencies by @eginhard in #50
- fix(xtts): load tokenizer file based on config as last resort by @eginhard in #57
- Skip audio files that can't be decoded by @eginhard in #66
- Fix wrong references in documentation by @hykilpikonna in #74
- fix(build): restrict spacy version to unbreak installation by @KoljaB in #92
- ci(release): switch to cibuildwheel by @eginhard in #95
New Contributors
- @ChristianRomberg made their first contribution in #45
- @abrahammathews2000 made their first contribution in #53
- @gravityrail made their first contribution in #59
- @akshatrocky made their first contribution in #64
- @hykilpikonna made their first contribution in #74
- @shavit made their first contribution in #85
- @KoljaB made their first contribution in #92
Full Changelog: v0.24.1...v0.24.2
v0.24.1
v0.24.0
What's Changed
Features
- Move from setup.py to pyproject.toml, simplify requirements by @eginhard in #19
This removes language-specific G2P dependencies for Bangla, Korean and Mandarin Chinese from the default install. They must now be installed via optional extras, as is already the case for Japanese: https://github.com/idiap/coqui-ai-TTS?tab=readme-ov-file#optional-dependencies
Fixes
- fix(espeak_wrapper): capture stderr separately by @eginhard in #17
- fix(bark): add missing argument for load_voice() by @eginhard in #22
- fix(server): ensure logging output gets actually shown by @eginhard in #26
Full Changelog: v0.23.1...v0.24.0