This is by no means an announcement for a project, but rather a discussion on AI voice deepfaking. I’ve seen and heard some really great audio Deepfake work done recently and it got me to thinking about the international cut scenes in The Good, The Bad, and The Ugly.

The performances are all there, but the voices are either aged too much, or in the case of Lee Van Cleef,replaced completely.

I’d like to try my hand at messing with this technology. I mean we have quite a lot of recordings of all three actors speaking from this era, so it should be feasible.

Has anyone else considered this?