This is the first and v0.0.9 release of TTS, an open text-to-speech engine. TTS is still an evolving project and any upcoming release might be significantly different and not backward compatible.
In this release, we provide the following models.
| Language | Dataset | Model Name | Model Type | Download |
|---|---|---|---|---|
| English | LJSpeech | TacotronDCA | tts | 💾 |
| English | LJSpeech | Glow-TTS | tts | 💾 |
| Spanish | M-AILabs | TacotronDDC | tts | 💾 |
| French | M_AILabs | TacotronDDC | tts | 💾 |
| English | LJSpeech | MB-MelGAN | vocoder | 💾 |
| Multi-Lang | LibriTTS | FullBand-MelGAN | vocoder | 💾 |
| Multi-Lang | LibriTTS | WaveGrad | vocoder | 💾 |
Notes
- Multi-Lang vocoder models are intended for non-English models.
- Vocoder models are independently trained from the tts models with possibly different sampling rates. Therefore, the performance is not optimal.
- All models are trained with phonemes generated by espeak back-end (not espeak-ng).
This release has been tested under Python 3.6, 3.7, and 3.8. It is strongly suggested to use conda to install the dependencies and set-up the runtime environment.