One photo, one track, one clip
No filming and no studio. The model animates the mouth region against the audio while the rest of the face stays as photographed.
What to upload
A front facing portrait with the mouth visible and a clean audio file. Sunglasses and extreme angles are the hard cases.
Where the voice comes from
Generate it with the narrator, clone your own with voice cloning, or upload a recording you already have.
FAQ
Do I need consent?
Yes, and it is enforced. Making a real person appear to say something they did not say is not acceptable use.
How long can it be?
Four to eight seconds per run. Split longer scripts into takes.
Does the whole face move?
Mostly the mouth region, with small natural movement. Expression stays as photographed.