OmniHuman 1.5
Audio-driven digital human — generate cinematic video from a single image and audio with emotional expressions
Inputs
Character ImageRequired
Upload a character image
JPG, PNG. Max 5MB, max 4096×4096. Higher resolution = better quality
Audio FileRequired
Upload an audio file (speech, song, etc.)
Max 60s (best under 15s). Supports speech, singing, and any audio
PromptOptional
0/300
Resolution
Higher quality (RTF ~27x)
How it works
- 1.Upload a character image (human, anime, pet — higher res = better quality)
- 2.Upload an audio file (speech, singing, etc. — under 15s recommended)
- 3.Optionally add a prompt to control camera, emotions, and actions
- 4.AI generates emotionally expressive video synced to your audio
Result
Your digital human video will appear here
Upload an image and audio to get started