Audio instead of text input to synthesize or vocode? (target audio prompt)

CorentinJ / Real-Time-Voice-Cloning

Clone a voice in 5 seconds to generate arbitrary speech in real-time

Other

51.65k stars 8.66k forks source link

Audio instead of text input to synthesize or vocode? (target audio prompt) #1169

Open remycoopermusic opened 1 year ago

remycoopermusic commented 1 year ago

Hi, I'm new to python and ML in general. I've got it to work on my mac m1, so that's nice. I've got the text to speech working but I was wondering: is it possible to learn a voice with a dataset and use that voice to replace a recorded voice (so not text)? Let's say I'm singing something but I want the voice of someone else. I replace my voice with the ai voice. Is that feature available? Does someone know?

remycoopermusic commented 1 year ago

after a bit more research; it seems to be vocoding only. Not synthesizing.