Releases: coqui-ai/TTS
v0.6.2
What's Changed
- Fix multilingual recipe by @Edresson in #1354
- Fix recipes as to the recent API changes. by @erogol in #1367
- Add docsqa to docs website by @nomagick in #1363
- REBASED: Add support for the speaker encoder training using torch spectrograms by @Edresson in #1348
- Add alphas to control language and speaker balancer by @Edresson in #1216
- Add Voice conversion inference support by @Edresson in #1337
- Update issue template by @erogol in #1370
- Open bible dataset formatter by @Edresson in #1365
- REBASED: Transform Speaker Encoder in a Generic Encoder and Implement Emotion Encoder training support by @Edresson in #1349
- Fix typo workflow text by @WeberJulian in #1403
- Add CITATION.cff by @erogol in #1404
- Fix default phonemizer for ja and zh by @WeberJulian in #1399
- Make Style by @erogol in #1405
- Fix #1380 by @erogol in #1409
- Hinge Gruut version to 2.2.3 by @erogol in #1419
- Update CheckSpectrograms notebook by @erogol in #1418
- Fix #1423 by @Edresson in #1424
- Update model file extension by @erogol in #1422
- Fix model manager by @erogol in #1436
- Add formatting tests by @WeberJulian in #1437
- Update base model wrt 👟 by @erogol in #1406
- Replace webrtcvad by silero-vad by @Edresson in #1431
- Bug fix in freeze encoder by @Edresson in #1391
- Enforce phonemizer definition for synthesis by @WeberJulian in #1441
- Fix G2P backend of the released models by @WeberJulian in #1461
- Add EmbeddingManager and BaseIDManager by @Edresson in #1374
- Update requirements coqui_trainer -> trainer by @erogol in #1478
- Update CONTRIBUTING.md, fix header by @Jackiexiao in #1463
- Add African models by @WeberJulian in #1511
- Print Model's license when downloading by @erogol in #1512
- Improve docsQA default questions by @nomagick in #1411
- patch print license by @WeberJulian in #1514
- v0.6.2 by @erogol in #1353
New Contributors
- @nomagick made their first contribution in #1363
- @Jackiexiao made their first contribution in #1463
Full Changelog: v0.6.1...v0.6.2
Contributors
Assets 2
v0.6.2 models
This release add 6 new VITS models for languages of the openbible dataset.
- ewe
- hausa
- lingala
- yoruba
- asante-twi
- akuapem-twi
Original work (audio and text) by Biblica available for free at www.biblica.com and open.bible.
v0.6.1 models
What's Changed
- Renamed all checkpoints from
model_file.pth.tartomodel_file.pth - Tested and fixed for all tts models the
"phonemizer"backend key in config
For best performance, you should use the commit version attached to each model
v0.6.1
v0.6.0
What's Changed
Tokenizer API
Tokenizer API is defined by the TTSTokenizer class. It is intended to provide all the text processing functionalities to a tts model. New tokenizers can also be added by subclassing the TTSTokenizer class.
Phonemizer API
Phonemizer API is defined by the BasePhonemizer class and implemented by the ESpeak and Gruut wrappers, ZH_CH,
JP_JA phonemizers. New phonemizers can be added by implementing the BasePhonemizer class.
BaseCharacters
BaseCharacters class provides an API to define the model vocabulary and provide the dictionary to map characters to
token IDs and back. There are two pre-defined classes inheriting from BaseCharacters. IPAPhonemes and Graphemes that respectively define the IPA phoneme character set for models using phonemes and grapheme set for models using raw characters.
Punctuations class
Punctuations class to strip out punctuations and restore them when needed.
Language specific text normalization routines under TTS.tts.utils.text
Under TTS.tts.utils.text there are folders for each language to accommodate the text normalization routines that
are designed for the language.
👟Trainer
We separate the trainer as a new repo 👟Trainer. It is a general-purpose model trainer for Pytorch with certain design choices in mind.
- Support for different experiment tracking dashboards like ClearML, Tensorboard, MLFlow, and W&Bs.
- Flexible to train any kind of DL model.
- Simple code base and easily expandable.
- Easy to debug.
It is a very early-stage and monolithic library currently. Feel free to share your ✨feedback✨ and ✨contribute✨.
VITS implementation update
With this version of VITS model, we get rid of some of the issues that affect the model performance. It also illustrates well how you could adapt any open-source model implementation to 🐸TTS and 👟Trainer without even knowing the rest for 🐸TTS library.
Full Changelog: v0.5.0...v0.6.0
New Models
-
GlowTTS + HifiGAN Turkish by 👑Fatih Akademi
$ tts --model_name tts_models/tr/common-voice/glow-tts --text "Bu bizim için oluşturulmuş bir örnek sevgili dostum." -
VITS and GlowTTS Italian by 👑@nicolalandro using MAI Italian male and female subsets.
Female VITS model
$ tts --model_name tts_models/it/mai_female/vits --text "Questo è un esempio per noi, mio <200b><200b>caro amico."Male VITS model
$ tts --model_name tts_models/it/mai_male/vits --text "Questo è un esempio per noi, mio <200b><200b>caro amico."
Contributors
Assets 2
v0.6.0 models
Merge pull request #1027 from coqui-ai/dev v0.5.0
v0.5.0
What's Changed
- Fix some setup papercuts by @reuben in #1022
- Add additional datasets by @loganhart420 in #1021
- Add UK vocoder models by @erogol in #1031
- Add multilingual models support by @erogol in #1007
- Implement YourTTS by @WeberJulian and @Edresson
- Fixes before YourTTS merge by @WeberJulian in #1044
- Fix language assignment by @erogol in #1047
- Fix if else statement by @erogol in #1050
- Fix train_tts.py and uncomment code by @WeberJulian in #1051
- v0.5.0 by @erogol in #1027
New Contributors
Full Changelog: v0.4.2...v0.5.0
v0.5.0_models
Model releases accompanying v0.5.0
v0.4.2
What's Changed
- Model zoo tests by @erogol in #900
- v0.4.2 by @erogol in #901
- Optional silence trimming during inference and find_endpoint() fix by @george-roussos in #898
- Update gruut to version 2.0 by @synesthesiam in #882
- Documentation corrections for finetuning and data preparation by @gullabi in #931
- server: fix compatibility with tts_models/en/ljspeech/fast_pitch by @Mic92 in #893
- v0.4.2 by @erogol in #914
New Contributors
- @george-roussos made their first contribution in #898
- @gullabi made their first contribution in #931
Full Changelog: v0.4.1...v0.4.2
