ByteDance OmniHuman 2.0 (Talking Video)
ByteDance OmniHuman 2.0 turns one picture and a voice track into a video of it talking, singing or performing, up to 5 minutes long in 720p.
OmniHuman 2.0 by ByteDance brings a single picture to life from an audio track: the person, cartoon character or animal in the image speaks, sings or performs in time with the sound, with matching lips, facial expressions, gestures and body movement. The video is as long as the audio, from 4 seconds up to 5 minutes, in 720p.
Use it for presenters and explainers, product and social videos, music clips and giving a voice to a character. Inputs: image (one picture, 300-6000 px per side, aspect ratio 0.4-2.5, under 30 MB), audio (WAV or MP3, 4-300 seconds, under 15 MB) and an optional prompt that directs the scene, the movements and the camera (Chinese, English, Japanese, Korean, Spanish or Indonesian).
It does not write or voice the speech: bring the audio, recorded or made with a text-to-speech model. The result has no watermark.
Price: $0.07 per second of video, counted in whole seconds; a 30-second clip costs $2.10. A refused or failed run costs nothing.
- Image to Video
Price: $0.07 per second of video, counted in whole seconds; the video is as long as the audio (4-300 s).
Run ByteDance OmniHuman 2.0 (Talking Video) via API
curl -X POST https://api.flymy.ai/api/v1/flymyai/bytedance-omnihuman-2_0/predict \
-H "X-API-KEY: $FLYMYAI_API_KEY" \
-H "Content-Type: application/json" \
-d '{"image":"https://storage.googleapis.com/open-inference-results/agent-files/99/ingests/954c188c570e5d1ab609546c0edfe481/14ba39b00b65d9bfe83550ebc5c1ccb0692a0658334c9a129e1e9ca544037ef5/omnihuman2-demo-presenter.jpg","audio":"https://storage.googleapis.com/open-inference-results/agent-files/99/8d8bdc561b92490c8d7a75c743b51ea8/tts_98a2092a072242a99e31833cd4f7e2fe.mp3","prompt":"He talks warmly to the camera, smiles and gestures with his hands, the camera slowly pushes in"}'API documentation · Create a free account · Pricing