Skip to main content
Percify turns a single photo into a video in two ways. To make the person in the photo speak, use Clone Yourself: add the photo and the audio, and Percify lip-syncs the face to it. To make the photo move without speech, run an image-to-video model in the Playground, where the price shows before you generate. The same photo can become a vlog, a UGC clip, a cinematic shot or an animated character:

Vlog

UGC

Cinematic

Animated

​
Make a photo talk

1

Open Clone Yourself

In the sidebar, click Clone Yourself and add a clear, front-facing photo, or take one with Use Camera.
2

Add the audio

Record yourself, upload an audio file, or pick a saved voice under My Voices.
3

Generate

Pick Fast or High, check Estimated cost, and press Generate video.
A talking video is charged by the length of the audio: The full walkthrough is in Create your AI twin with Clone Yourself. To make a face in an existing video speak new audio, see Lip-sync a video to any audio.

​
What it costs

A talking video is priced on the length of the audio, at 2 to 6 credits per second depending on the quality you pick, so a 10 second clip costs 20 credits in Fast mode. An image-to-video model is priced per run instead, from 2 credits for the cheapest model to 240 for the most expensive. The price for your settings always shows before you generate.

​
Animate a photo without speech

Open the Playground, choose an image-to-video model, add your photo and a prompt that describes the motion, and run it. Each model has its own lengths and price, listed in AI models in Percify. For several shots from one prompt, see Make a multi-shot video in seconds.

Make your first talking video

The shortest path from a photo to a finished video

Write prompts for AI video

Describe motion so the model follows it

Lip-sync looks off

Fixes for talking videos that look wrong

How Percify credits work

What each action costs
Last modified on September 19, 2026