OmniHuman 2.0 is an audio-driven AI video generation model from ByteDance, designed to turn a single image and an audio track into an animated character video.

Users can create videos of characters speaking, singing, and moving with synchronized lip movements, facial expressions, and body gestures. The platform supports videos up to 5 minutes, with optional text prompts to guide character actions and camera movements. OmniHuman 2.0 can be used by content creators, educators, marketers, animators, storytellers, and creative teams.

Creators can make talking character videos and music performances. Educators can turn recorded lessons into presenter-style videos, while marketers can develop product presentations and promotional concepts.