tango icon indicating copy to clipboard operation
tango copied to clipboard
verified publisher icon declare-lab

Producing audio in different Sample Rate

Open cvillela opened this issue 2 years ago • 3 comments

Hey!

I was wondering if it was possible to train the model in 48kHz audio, and then generate audio directly in 48kHz. Has anyone attempted this?

cvillela avatar Jul 10 '23 21:07 cvillela

That is definitely possible and would be really great to have! We could not try this due to computational constraints.

deepanwayx avatar Jul 15 '23 03:07 deepanwayx

Awesome! Will try it out. How much VRAM do you think is necessary for attempting it?

cvillela avatar Jul 21 '23 14:07 cvillela

@deepanwayx Also, I see that the "Tango Prompt Bank" is all in 16.000Hz. Would you guys have the raw dataset, not resampled, available?

cvillela avatar Jul 21 '23 17:07 cvillela