OmniHuman 2.0 - Turn one image and audio into a 5-minute avatar video

by•
OmniHuman 2.0 turns one image and an audio track into a 720p talking, singing, or acting character video up to five minutes long. Audio drives lip sync, emotion, and performance, while optional prompts control actions, camera, and background. It supports multi-person scenes, pets, cartoons, and pixel art, with 20 free credits to start.

Add a comment

Replies

Best
Maker
📌
Hi Product Hunt! We built OmniHuman 2.0 for creators who need expressive avatar videos without filming. Upload one image and 4–300 seconds of audio, then optionally describe the action or camera. We focused on long-form consistency, natural lip sync, and practical creative control. I’d love to hear what you would make with it and where the workflow could be better.