मुख्य सामग्रीवर जा

A real video · a WAV, MP3 or M4A sound · up to 600 seconds a run

AI Lip Sync: Make the Lips in a Real Video Follow a New Sound

You have the video and you have the words, just not together: a line you fluffed and recorded again, a talk you dubbed into Hindi, a take where the sound was better than the picture. Face Studio’s Lip sync takes a real video of a person and a sound file and makes the mouth move with the new audio. It does not invent a voice and it does not animate photos — you bring the video and the sound, it brings them into step.

  • Finding people is free
  • Up to 8 people, you pick who
  • AI-edited label on every result
Meera4 of 10 photos
Kabir6 of 10 photos
  • Swap
  • Enhance
  • Age
  • Expression
  • Lip sync

Meera → Person 2 · Kabir → Person 1

Illustration with fictional characters: pairing faces to people, as in Face Studio. Not a swap result.
  • SwapYour face onto the person you pick
  • EnhanceSharpen a soft or low-light face
  • AgeYounger or older, one slider
  • Expression9 sliders: smile, brows, eyes
  • Lip syncA video says the audio you give it

What AI lip sync does — and what it doesn’t

Lip sync is one of Face Studio’s face edits. “Lip sync” means different things on different sites, so here is exactly what this one does.

  • It works on a real video

    Give it a clip of a person talking or singing. A photo is refused: Lip sync does not make still pictures talk.

  • You bring the sound

    A WAV, MP3 or M4A file — a line you recorded again, a dub, a voice-over. Face Studio does not generate or clone a voice.

  • The mouth follows the sound

    The lips are made to match the audio, which plays from the first frame of the video. Everything else is the video you uploaded.

  • One face or everyone

    Face Studio finds up to 8 people in the clip. Pick the one who is speaking, or leave nobody picked and every face follows the sound.

How to lip sync a video to new audio

Four steps in Face Studio, on a phone or a computer.

  1. Choose Lip sync and add the video

    Upload a clip, or paste a link from YouTube, TikTok, Instagram, Facebook, Vimeo, X, Twitch or Dailymotion to a video you made or have permission to use.

  2. Pick the speaker

    Tap the person whose lips should move. With nobody picked, every face in the frame follows the sound.

  3. Add the sound

    Choose a WAV, MP3 or M4A file. It plays from the start of the video, so trim it to begin where the speech should begin.

  4. Sync and compare

    Press Sync the lips, watch the result with the before-and-after slider, then download or share it.

Getting a good sync

Lip sync redraws a mouth to match a sound, so the clearer both are, the better it looks. Unlike the other face edits, it has no one-frame preview — a single frame cannot show movement — so a short clip is the cheapest test.

  • A mouth you can see

    The speaker facing the camera, with no hand, microphone or deep shadow over the mouth.

  • Clear speech

    A voice without loud music or crowd noise under it is the easiest kind of sound to follow.

  • Sound that starts on time

    The audio plays from the first frame, so start it there and keep it about as long as the clip.

  • Short before long

    Try a short clip first. Each run covers up to 600 seconds; a longer video is offered in parts.

Uses that work well

Lip sync is at its best when the person in the video is you, or someone who agreed to it.

  • Fix a line without reshooting

    Record the sentence you got wrong again and sync your own video to it.

  • Dub your own videos

    Record a dub of your talk, tutorial or product demo in another language — Hindi, say — and sync your video to it.

  • Use the better take

    When one take had the right sound and another the right picture, put them together.

  • Perform your own song

    Sync a video of you singing to the final mix of your own song or cover.

Never put words in someone else’s mouth

Lip sync changes what a person appears to say, which is why its rules are strict. Face Studio’s terms, which you accept before your first swap or edit, allow your own face, an adult who gave you permission, or a face that is not a real person — and forbid using an edit to impersonate someone in order to deceive, mislead or defraud.

Every result carries an invisible watermark and a C2PA content credential saying it was edited with AI. Anyone can report a result, results that break the terms are taken down, and faces that appear to be minors are refused.

Questions

It remakes the mouth in a video so it moves with a different audio track. In Face Studio you give it a real video and a sound file, and the lips follow the sound.

Pick the face. Pick the person. Check it before you commit.

Sign in to start. Finding the people in a video is free.