MelGAN vocoder - Githubissues

KevinMIN95 / StyleSpeech

Official implementation of Meta-StyleSpeech and StyleSpeech

MIT License

241 stars 39 forks source link

MelGAN vocoder #2

Closed vsgogoryan closed 3 years ago

vsgogoryan commented 3 years ago

As I understand, you train your own version of MelGAN for multi-speaker synthesis, as the official code supports the sampling rate of 22.05 kHz, while StyleSpeech operates at 16 kHz. Could you share the details for reproducibility purposes: which dataset did you use, which parameters did you change? Or you can maybe upload the trained vocoder itself? It would be great!

KevinMIN95 commented 3 years ago

Sorry for I can not share the trained vocoder. But I use the same LibriTTS dataset for training the MelGAN and I didn't change any parameters from the default ones.

vsgogoryan commented 3 years ago

Got it, thank you!