PlayVoice / Grad-SVC

Diffusion Singing Voice Conversion based on Grad-TTS from HuaWei
https://huggingface.co/spaces/maxmax20160403/grad-svc
MIT License
123 stars 15 forks source link

What is the advantage for Grad-SVC, compare to So-VITS-SVC? #2

Open MuruganR96 opened 1 year ago

MuruganR96 commented 1 year ago

Hi @MaxMax2016 Thank you for this wonderful project

What is the advantage for Grad-SVC, compare to So-VITS-SVC?

MaxMax2016 commented 1 year ago

Grad-SVC is a diffusion based SVC. So ~ It has no electronic sound, but is sensitive to noise data.

MuruganR96 commented 1 year ago

How much time of data would be good to build a model? Please help me. I need your guidance.

MaxMax2016 commented 1 year ago

It's not perfect yet. If you are not a developer, you should wait a while until the perfect state is released