Audio Driven - Video Input instead of image

OpenTalker / StyleHEAT

[ECCV 2022] StyleHEAT: A framework for high-resolution editable talking face generation

MIT License

636 stars 78 forks source link

Audio Driven - Video Input instead of image #44

Open oijoijcoiejoijce opened 1 year ago

oijoijcoiejoijce commented 1 year ago

Thanks for the great work; do you know how we can run audio-driven reenactment but have the input be a video instead of an image? (i.e. sync the lips in that video)

Hujiazeng commented 1 year ago

Hi, were you able to solve this issue?