Lip Sync takes a video with a face and an unrelated audio file, and re-renders the mouth shapes so they match the new audio perfectly.
Use cases
- You re-recorded narration and need the face to match the new audio.
- You want a real person's face to deliver a script they didn't record.
- You translated audio and the original mouth shapes don't match the new language.
Precision vs Speed
Precision mode uses HeyGen's avatar-grade lip-sync model. Higher quality, ~2× slower, ~2× the credits. Use for hero content.
Speed mode is for drafts and batch work. Quality is still high — for content that won't be watched on a big screen, it's often indistinguishable.
Audio prep
The cleaner the audio, the better the sync. Run Voice Isolator on your audio first if there's background noise. Aim for a tight, dry voice track.
Captions option
Lumen can burn captions from the new audio directly into the synced video. Good for accessibility and for sound-off feeds.