AI talking photo
Make a photo speak — one you have the rights to.
Upload a clear, front-facing photo, add a line or an audio track, and get a short talking video. The face is driven directly, not redrawn.
What works
- One clear, front-facing face
- Mouth visible and unobstructed
- Even light, no heavy motion blur
What this does not do
- It does not translate, and it speaks English voices only
- It does not redraw the face — the photo you upload is what moves
- It does not check whose face this is. You confirm the rights before generating.
A human face works best. For a pet, use AI Pet Podcast — it is built for animals.
2. What should they say?
Performance
See the difference between Quick and Vivid
Same photo, same line. Both clips are silent on purpose — press play if they do not start.
Write a line or add audio above and the exact cost appears here.
Questions
Before you upload
Whose photo can I use?
Your own, or someone's whose permission you have. Public figures, politicians and celebrities are not allowed, and neither is anyone under 18. You confirm this before generating; we do not check it for you.
Do you check the photo?
No. There is no face recognition and no age detection here. The three confirmations you tick are the record of what you told us, and the acceptable use policy is what we enforce on report.
What makes a photo work?
One clear, front-facing face with the mouth visible and unobstructed. Sunglasses are fine; a hand or microphone across the mouth is not.
How long can the result be?
There is no 8-second plan limit. Quick and Vivid accept 5–60 seconds of uploaded or recorded audio; Any subject accepts up to 15 seconds. Cost follows the generated speech duration and your available credits.
Is this the same as the pet tool?
The underlying animation is the same. Pet Podcast adds a studio scene around your pet first; this does not — a photo of a person is driven directly, so the face is not redrawn.