AI Lip Sync

AI Lip Sync turns one portrait photo and one audio clip into a video in which the person speaks the audio, with matching mouth and head movement.

Make a talking video

1. Add a portrait: JPG, PNG or WebP, up to 20 MB. Use a clear, front-facing face or upper body with the mouth visible.
2. Add the audio: MP3, WAV, M4A or AAC, up to 25 MB and 30 seconds. Make it with the Voice Generator if you don’t have a recording.
3. Choose 720p or 1080p and press Generate.

Photos that work best

• One person, facing the camera, with the whole face in view.
• Avoid side profiles, full-body shots, blurry photos, or hands and objects in front of the mouth.
• Some photos of real people may be refused by the model’s safety filter. An AI-made portrait or a Character works reliably.

What it costs

Charged per second of audio; 1080p costs more than 720p. The video is saved to My Creations.