Real-Time-Voice-Cloning icon indicating copy to clipboard operation
Real-Time-Voice-Cloning copied to clipboard

Published 20 hours ago • verified publisher icon CorentinJ

Results of training on my own dataset

Open dongdongdashen opened this issue 3 years ago • 1 comments

Hi,When I train with my own dataset, the result is as shown in the following picture.Can you tell me where is the problem? May be the number of data is too small? (my dataset includes 20 minutes of voice data in total ) Looking forward to your reply : step-5000-align step-5000-mel-spectrogram )

dongdongdashen avatar Feb 28 '22 08:02 dongdongdashen

Yes 20 minutes is nt enough! you can use this much data if you fine tune on pre trained weights!

ireneb612 avatar Mar 01 '22 14:03 ireneb612