Uberduck

it might be because you have very short wav files

but it might also be because of batch size

I put the batch size to 1 and that was the result

https://github.com/Rongjiehuang/GenerSpeech

Like individual vocalist seperation tools? Or just vocals from insrtuemnts?

I never did but I've been searching for something like that a VERY long time

I found asteroid but it doesn't work great

Give me the link im getting pictures of asteroids

<@1018356568095461388>
It works okay on singing, but even better on speech
Only major issue i've found is that sometimes it doesn't separate at all and also the pretrained models have a sample rate of either 8 or 16 khz https://github.com/asteroid-team/asteroid

I want to update the voice of merasmus and bombinomicon, someone asks me a favor about how I train the voices with the new and better tacotron2,  pipeline Tacotron 2 model, since the legacy seems somewhat obsolete to me

new chatbot dropped https://whisper.ggerganov.com/talk/