Source image identity
Start from one portrait or character image and preserve the recognizable visual subject.
Image and Audio Performance
LipSync
Turn a source image and audio into a short performance video. LipSync is built for talking head tests, character performance drafts, singing previews, and audio-led visual experiments.
Quick start
How to use LipSync
- Upload a clear source image. Use a portrait or character frame where the face is visible and the identity should be preserved.
- Add speech or singing audio. Choose the audio segment that should drive the mouth movement and performance timing.
- Generate and review the clip. Compare the result against the audio, facial identity, framing, and intended performance style.
Best for: talking head clips, character voice tests, singing previews, audio-led social video drafts, and production review.
Audio-led character performance from one image.
LipSync gives creators a focused way to pair a still image with audio, producing a short generated video that follows speech or singing while keeping the source character recognizable.
Audio-timed motion
Use speech or singing audio to guide mouth movement, rhythm, and performance pacing.
Short review clips
Create compact performance drafts for voice tests, character studies, and creative iteration.