audio examples
easypx opened this issue · 1 comments
https://vocaroo.com/1avUJ4T2Dgr8
Source: Dave Gahan Acapella
Singer: configs/singers/singer0015.npy
Model: maxgan_pretrain_32K.pth
Can you tell what gender the singers in maxgan_pretrain_32K.pth were? I assume they are mixed men and women. 🙂
yes, they are mixed men and women. maxgan_pretrain_32K.pth is trained by the data: https://github.com/Multi-Singer/Multi-Singer.github.io
Thx, I'll take a look. (I've already figured out a lot by transposing.)
Here is the full mix of "Enjoy The Silence"! https://www.youtube.com/watch?v=3Y4UjhM0A6A
ABBA - The Winner Takes It All (fully AI synthesized voice/PoC)
https://www.youtube.com/shorts/ou0atcGNXrs
Synthesized voice to custom model voice cloning. Model training took 3 nights (~30h in CPU mode/5000 iterations).
I used ~10 min of clean English singing data with maxgan_pretrain_32K.pth, learning rate 1e-5.
Base-voice: Synthesizer V Studio Pro (Solaria), extracted
Comparison: https://www.youtube.com/watch?v=KX0lcVr9ugI