Higgs Avatar API is now available
The Higgs Avatar generation API is now live for creating talking-head videos.
Audio-to-video
Upload a still image and drive the avatar with an audio clip.
Text-to-video
Pass text through input_tts to synthesize speech with Higgs TTS and animate the avatar in the same request.
How it works
- 01Create a video job with
POST /v1/videos. - 02Poll
GET /v1/videos/{video_id}until the job is complete. - 03Download the rendered MP4 from
GET /v1/videos/{video_id}/content.
Release notes
This release is the avatar generation API: it is built for creating videos from image plus audio or text. Use POST /v1/videos for asynchronous jobs, or POST /v1/videos/stream when you want the streaming version and playback can begin before the full clip is ready.
Avatar chat is a separate realtime experience, and will be released later together with Higgs Realtime.
Start with the API docs
Acknowledgments
Technical Contributors: Huapeng Zhou, Xinyu Wang, Junming Chen, Erik Li, Jixuan Wang, Weisu Yin, Yizhi Liu, Mu Li