kan-bayashi / ParallelWaveGAN

Unofficial Parallel WaveGAN (+ MelGAN & Multi-band MelGAN & HiFi-GAN & StyleMelGAN) with Pytorch
https://kan-bayashi.github.io/ParallelWaveGAN/
MIT License
1.54k stars 339 forks source link

add discrete vocoder #427

Open wyh2000 opened 4 months ago

wyh2000 commented 4 months ago

We add a HifiGAN based vocoder which can decode hubert tokens to waveform. It supports LJSpeech now. It is aligned with configurations in discrete TTS from ESPnet.