Yeah, for sure! For Ashton's Photoshoot I'll probably use TTS just in case there is an accent TF in her future, but I think voice-to-voice would be very viable. Possible problems would be a) mic selection/placement to get good audio quality while keeping the mic out of the shot and b) giggling/laughing which might cause the AI to glitch. I will likely test this type of thing more in the future though.
Blankage
2025-09-01 03:47:13 +0000 UTC
Rather than lipsync, couldn't you just use your actual speech and do voice-to-voice clone? That way the lipsync is automatic and perfect, and IMO the audio tends to be better than TTS too.
fizzleus
2025-09-01 03:41:34 +0000 UTC
It's a custom-trained voice clone in ElevenLabs
Blankage
2025-08-31 23:51:16 +0000 UTC
Pretty solid What voice did you use for the sync test ?