# Working notes on lip sync vs mimic motion

If you only read one section, read the part about choosing the source image; it decides most of the result.

A note I keep at the top of this repo so I pick the right Fameo tool.

| If the viewer should notice… | Use |
| --- | --- |
| A line of speech or a song | [the Fameo generator](https://fameo.ai/lip-sync-generators/ai-lip-sync-generator) (or the pet, cartoon, drawing, singing, or multi-talk pages) |
| A wave, a dance step, a gesture | [mimic-motion tool](https://fameo.ai/ai-mimic-motion) |

Mimic motion takes a photo plus a motion video (MP4 or MOV, up to 120 seconds and under 20 MB) or a preset. You can keep the photo’s background or use the video’s, and render at 480p or 720p. A clear, single, front-facing full or half-body shot works better than a tight head crop.

Lip sync takes an image plus text or audio and returns a talking video in minutes.

Both require authorized media. I do not copy a performer’s movement or a person’s face without permission. Start at [make a photo talk](https://fameo.ai).

Before posting, watch the clip once with the sound off. If the mouth still reads as speech, it is ready.