How to train HiFi-GAN on the VCTK dataset as its wavs is the style of ‘.flac’

jik876 / hifi-gan

HiFi-GAN: Generative Adversarial Networks for Efficient and High Fidelity Speech Synthesis

MIT License

1.91k stars 504 forks source link

Open DthdZK opened 11 months ago

Ziyi6 commented 4 months ago

I think simply write a function which can read the flac format replacing the origin one, and that's it