News for hackers. Stuff that matters.

A comment on Microsoft’s new text-to-speech model can duplicate anyone's voice in 3 seconds. Parent comment

Re:Microsoft’s new text-to-speech model can duplicate anyone... by deegles on Monday January 09, 2023 @09:50AM (#34310922)
I can think of one... podcast cleanup app:

1. Speech to text

2.fix up/edit text with GPT-3

3.text to speech in the original speaker's voice(s), preserving prosody and inflection with Vall-e.

If done with every participant's consent I don't see how it's not legitimate.

[ Parent | on HN ]
Re:Microsoft’s new text-to-speech model can duplicate anyone... by burlesona on Monday January 09, 2023 @09:53AM (#34310964)
Descript already does this.
[ Parent | on HN ]
Re:Microsoft’s new text-to-speech model can duplicate anyone... by morrisjm on Monday January 09, 2023 @11:55AM (#34312512)
you could also just skip step 1 https://twitter.com/lexman_ai
[ Parent | on HN ]