Audiovisual-Synthesis icon indicating copy to clipboard operation
Audiovisual-Synthesis copied to clipboard

Published 20 hours ago • verified publisher icon dunbar12138

Data required for Training

Open Anchit1999 opened this issue 4 years ago • 0 comments

To train a model from scratch, it needs about 30 minutes of the target speaker’s speech data and around 10k iterations to converge Is it a single 30 minute audio file or can be multiple small audio files?

Anchit1999 avatar Jun 21 '20 18:06 Anchit1999