search

coqui-ai / TTS

🐸💬 - a deep learning toolkit for Text-to-Speech, battle-tested in research and production
http://coqui.ai
Mozilla Public License 2.0
35.8k stars 4.38k forks source link

Representative Samples #409

Closed Dmole closed 3 years ago

Dmole commented 3 years ago

The README.md links to English Voice Samples which claim to use and English DDC model however it is not identifiable in list models;

tts --list_models
 Name format: type/language/dataset/model
 >: tts_models/en/ek1/tacotron2
 >: tts_models/en/ljspeech/glow-tts
 >: tts_models/en/ljspeech/tacotron2-DCA
 >: tts_models/en/ljspeech/speedy-speech-wn
 >: tts_models/es/mai/tacotron2-DDC
 >: tts_models/fr/mai/tacotron2-DDC
 >: tts_models/zh-CN/baker/tacotron2-DDC-GST
 >: tts_models/nl/mai/tacotron2-DDC
 >: tts_models/ru/ruslan/tacotron2-DDC
 >: vocoder_models/universal/libri-tts/wavegrad
 >: vocoder_models/universal/libri-tts/fullband-melgan
 >: vocoder_models/en/ek1/wavegrad
 >: vocoder_models/en/ljspeech/multiband-melgan
 >: vocoder_models/nl/mai/parallel-wavegan

Can the Samples page be changed to one in this project and using an available model?

erogol commented 3 years ago

You are right it is a bit confusing but we don't have such a plan anytime soon.