LatentSync

A model that redoes the lip movement in a finished clip to match a new audio track. You bring a video and a sound, and the person says exactly those words. The distinction is price: it stays the same up to 40 seconds of clip, so on short ad clips LatentSync comes out cheaper.

  • No subscription
  • Price shown before you run
  • Refund if a generation fails

Fit lips with LatentSync

How it works

  1. Open the “Voiceover” tab and pick the “Lips” mode.
  2. Pick a clip where the speaker is visible and a ready track: a voiceover from your feed or your own file.
  3. Choose the LatentSync model.
  4. Check the price and run it. The clip with new lips appears in your feed.

Specifications

What it doesFits lips to new audio
InputA clip and a ready track
Price up to 40 seconds$0.26 for all
A 60-second clip$0.39
A 120-second clip$0.78
Strong sideCheap on short clips

Good to know

One price up to 40 seconds

A 5-second clip and a 40-second clip cost the same, $0.26. For short ad clips this beats billing by the second.

By the second after that

Past 40 seconds the price grows by about $0.0065 a second: a 60-second clip costs $0.39, a 120-second clip $0.78.

For dubbing and translation

When the voice was swapped for another language, the lips still follow the old speech. This model fits them to the new track, and the clip stops looking dubbed.

VEED Lipsync alongside

If the face in the frame is large and you need care, look at VEED Lipsync: it holds the face more exactly, but it is billed by the second and costs more on long clips.

Price

Up to 40 seconds of clip the price is one: $0.26. After that about $0.0065 a second: a 60-second clip costs $0.39, 120 seconds $0.78. The price is shown before you run; if a generation fails, the money is returned.

FAQ

When should I pick LatentSync and when VEED Lipsync?

LatentSync is better on short clips up to 40 seconds: the price is fixed. VEED Lipsync holds the face more carefully, but it is billed by the second.

How much does a 15-second clip cost?

$0.26. A 40-second clip costs the same.

Where do I get the sound?

Make a voiceover in the “Voice only” mode, for example with ElevenLabs Turbo, or upload your own track: a voice-recorder file will do.

Will the result be flawless?

The result depends on the shot: sharp, frontal faces work best. If you need precision on a close-up, compare with the other option on a short fragment.

See also

Home · Ready-made presets