All apps
A tool in AI Video Creator

AI lip sync: new words, same take

Give a clip of your character new speech. The mouth is re-timed to every word, so you can change the line, fix a script or speak another language without filming again.

Every video on this page was made in SmileToAI.

Malcolm in the original clip

10 s clip · 9 s of new speech · lip sync 3,150 Sparks

Clips up to 2 minmp4, mov or webm
Up to 60 sof new speech per run
Any languagethe speech is in
From 350Sparks a second of speech
Change the words, keep the take

One clip of Nadia. Two languages.

The same ten-second clip, synced once to an English voiceover and once to a German one. Nothing was filmed twice.

English7 s of speech · 2,450 Sparks

Deutsch7 s of speech · 2,450 Sparks

Before and after

Each clip was made in the video studio from a ready-made character, then given a new line. Tap a pair to choose it for the player above.

Malcolm · Private chef

“Tonight's special is slow-braised short rib with a red wine glaze and creamy polenta. It's been in the oven since nine this morning.”

10 s clip · 9 s of new speech · lip sync 3,150 Sparks

Nadia · Bookkeeping firm

“Tax season doesn't have to be stressful. Send us your receipts by the fifteenth, and we'll handle the rest.”

10 s clip · 7 s of new speech · lip sync 2,450 Sparks

Bruno · Bistro mascot

“Pizza night is back! Fresh dough, our own tomato sauce, and as much cheese as you dare. Every Friday from six.”

10 s clip · 8 s of new speech · lip sync 2,800 Sparks

How it works

  1. 01

    Pick the character

    A ready-made one, one you generate in Elements, or you with your consent recorded on camera.

  2. 02

    Add a clip and the new speech

    A clip of that character, up to two minutes, and up to 60 seconds of speech from Voice Studio, a recording or an upload.

  3. 03

    Run it

    The price shows first. The synced clip lands in your Library, next to the original.

Only your character’s face

Lip sync checks the clip before anything is charged, and marks every result as AI.

  • The face must match

    The clip is compared with your character’s photos before the run. A clip of someone else is refused.

  • Nobody else in frame

    A clip that also shows another person is refused, so only your character’s mouth is changed.

  • Marked as AI

    Every synced clip carries a small AI mark and signed content credentials.

  • AI-made, or you with consent

    Use a character made by AI, or yourself after you record your consent on camera.

What 15 seconds of lip sync costs

You pay for the seconds of new speech, not the length of the clip. Face tools are part of paid plans.

See plans and Sparks
15 s of speech5,250Sparks · 350 a second

Questions

What does AI lip sync do?

It re-times the mouth in a clip of your character to new speech, so the clip says the new words. The rest of the take stays as it was.

Which clips work?

A clip of your character alone, up to two minutes, in mp4, mov or webm. A clip that shows anyone else is refused before anything is charged.

Can it change the language?

Yes. Give it speech in any language and the mouth follows those words, as in the English and German clips of Nadia above.

What if the speech is longer than the clip?

The clip is extended to fit the speech, and the run sheet tells you by how many seconds before you run it.

What does it cost?

350 Sparks a second of new speech, shown before you run it. Face tools are part of paid plans.

Where do I get a clip of my character?

Make one in AI Video Creator from the character’s photo, as these examples were, or use a recording of yourself after you record your consent.

Talking avatarNo clip? Make one talk from a single photo.Voice StudioWrite the new line and make the voiceover.AI Video CreatorTurn a character’s photo into the clip.

Give your next take new words

Lip sync a clip