search

TensorSpeech / TensorFlowTTS

:stuck_out_tongue_closed_eyes: TensorFlowTTS: Real-Time State-of-the-art Speech Synthesis for Tensorflow 2 (supported including English, French, Korean, Chinese, German and Easy to adapt for other languages)
https://tensorspeech.github.io/TensorFlowTTS/
Apache License 2.0
3.8k stars 810 forks source link

training fastspeech2 from scratch on a new dataset #701

Closed MostafaAlaviyan closed 2 years ago

MostafaAlaviyan commented 2 years ago

hi thanks for the valuable repository. I want to train fastspeech2 from scratch on a Persian dataset, but it is expressed that ''Right now we only support ljspeech, kss, baker, libritts, thorsten and synpaflex for dataset argument. In the future, we intend to support more datasets.'' on the main page of the repo. now, how can I train the model on my dataset? best regards @ZDisket

ZDisket commented 2 years ago

@MostafaAlaviyan It only supports those datasets out of the box, but you can add your own by analyzing how they're implemented and adding your own. I trained on a Spanish dataset, it's easy.

MostafaAlaviyan commented 2 years ago

thanks a lot, can u explain the way of defining a new dataset to the main repo clearly? where do I add the dataset? and how to add it? @ZDisket

v-nhandt21 commented 2 years ago

Hi @ZDisket , do you know any publish Spanish dataset for training TTS?