> ## Documentation Index
> Fetch the complete documentation index at: https://docs.zerogen.com/llms.txt
> Use this file to discover all available pages before exploring further.

# sync-3 lip sync

> Make a person in a video or photo speak along with an audio clip.

sync-3, from sync., matches the mouth in a video or a still image to an audio
clip, such as a line of dialogue. Choose it in the video generator.

## What you give it

Add two references:

* **A video** (MP4 or MOV, up to 30 fps) or **a still image** (JPEG or PNG) of
  the person.
* **An audio clip** (MP3, WAV, M4A, OGG or FLAC) from half a second to 10
  minutes long.

## Output settings

The result is an MP4 at the same frame rate as your video. A still image
becomes a 25 fps video.

When the video and audio are different lengths, **Length mismatch** chooses
what happens:

* **Bounce video to the audio**: plays the video forward and back. The default.
* **Loop video to the audio**
* **Stop at the shorter clip**
* **Retime video to the audio**: speeds up or slows down the video.
* **Keep the whole video, silent past the audio**

## Cost

sync-3 is priced per frame of the result, so a longer result, or one at a
higher frame rate, costs more. A 5-second result at 25 fps, such as one made
from a still image, costs about \$0.83. Generate shows the estimate once both
references are added. Prices are as of October 28, 2026.

Make a speech clip to lip-sync with [Eleven v4](https://docs.zerogen.com/models/eleven-v4.md).


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.