Create a lip-sync video

Lip sync creates a new video by matching dialogue audio to the visible speaker in a source video. The original video remains unchanged.

Prepare the source video

Choose footage with a visible, unobstructed face and a clear view of the mouth. Stable framing and natural head movement generally give the action a better foundation than fast cuts, extreme angles, or a very small face.

The source can come from a connected video node, existing brand media, or a supported upload. Confirm that you have permission to use the person's likeness and the source footage.

Prepare dialogue audio

Use clear speech with minimal background noise, music, echo, or overlapping voices. You can connect an audio node, choose saved audio, upload a supported file, or generate dialogue audio first.

Review pronunciation, timing, and performance before lip sync. The action synchronizes to the audio you provide; it does not repair a weak script or noisy recording.

Select a compatible model

Lip-sync choices are drawn from video operations that accept both video and audio inputs. The picker can show estimated credits and whether an option is premium, locked, or unavailable. Use the current options rather than relying on a fixed provider list.

Generate and review

Connect both required inputs, check the selected model and estimate, then generate. Review the mouth shapes, timing, face stability, audio alignment, and the transitions around pauses. Also compare the result with the original to catch identity or background drift.

If the browser connection closes, the run may still finish. Check the Lip Sync node before retrying. A duplicate retry can create another charge or result.

For dialogue creation, read Generate audio.