You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hello. I’ve trained a MeloTTS from scratch on the English dataset (demo test with 20 minutes). And it worked! So, is it architecturally possible to train it on raw singing and make a model to sing? If not, please tell me why?
Thank you!
The text was updated successfully, but these errors were encountered:
Hello. I’ve trained a MeloTTS from scratch on the English dataset (demo test with 20 minutes). And it worked! So, is it architecturally possible to train it on raw singing and make a model to sing? If not, please tell me why?
Thank you!
The text was updated successfully, but these errors were encountered: