fal-ai/flashtalk
URL of the face reference image.
URL of the speech audio file (WAV, MP3, etc).
Run the model to see the result here.