Existing video, replacement audio

Replace a video voiceover and match the mouth to it.

LipSync is for the point after you already have both inputs: an existing talking video and a new, authorized voice track that should appear to be spoken on camera.

Try the free preview

Bring an audio track you own. No account required for the free preview.

Hey, quick update — the new build ships on Friday, so keep an eye out.
Replacement audio selectedAI-generated presenter

What the sample proves

The audio drives the new mouth movement.

Both versions use the same presenter footage. Switch the line to see why replacement audio should be close in duration and clearly spoken.

Three steps

Bring the inputs. Keep control of the words.

01

Choose a talking-head clip

Use an MP4 or MOV with one visible, evenly lit face.

02

Bring the replacement audio

Use clean speech without music or overlapping voices; MP3, WAV, and M4A are supported.

03

Create a lip-synced MP4

The free preview accepts up to 8 seconds of replacement audio.

A good fit

Use it when

  • An approved revised voice track
  • Clear speech with no background music
  • A source video at least as long as the new audio

Keep in mind

Not the right tool for

  • Generating, cloning, cleaning, or translating audio
  • Music videos, animals, or heavily hidden faces
  • A source clip shorter than the needed performance

Use cases

Find the right starting point

Every workflow starts with the same two inputs: one clear speaker and audio you are authorized to use.