You must log in or register to comment.
I don’t know if it’s a mistake, but the various voices all sound identical.
FTA:
https://huggingface.co/hexgrad/Kokoro-82M/blob/main/SAMPLES.mdPossible as they only mentioned that new voices sound ‘much more natural’. Nothing about each sounds are unique or something.
Maybe just 1 voice available by default.
Which surely means there’s a neural network involved.
A technology we can trust everyone to be normal about.
They had natural sounding TTS a decade ago.
Ehhh.
