So far as I'm aware, that's not how the software works. Just as deepfake needs a face and a performance to overlay on top of, respeecher needs a vocal performance to act as the foundation of the generated overlay. Computers can't act, or dynamically inflect or emote to anything close to that degree.It doesn't sound like they used anyone for the basis of the Vader audio based on the article I linked above.
It was pure AI based off archived recordings, JEJ didn't even record anything new.
The article just didn't mention who did the actual performance, only that Jones consulted and advised. It could have been Hayden, it could have been Matt Wood, could have been Scott Lawrence, it could have been literally anyone with some acting ability, but I think it did indeed have to be *someone*.
Plus as has been pointed out, the point of the article isn't who did Vader's "voice", it's how the final effect was achieved by a small start-up company right in the midst of an invasion of their country.
Last edited: