A real-time interactive streaming digital human engine

A real-time interactive streaming digital human engine enabling synchronized audio-video conversation, widely adopted in commercial applications. LiveTalking leverages real-time streaming digital human technology to drive virtual avatars via text or voice, combined with LLM for intelligent conversation. Suitable for Virtual Streamer, AI Digital Human Customer Service, Intelligent Voice Assistant, Large Screen Presentation

Features

1. Supports multiple digital human models: ernerf, musetalk, wav2lip, Ultralight-Digital-Human
2. Supports voice cloning
3. Supports interrupting the digital human while speaking
4. Supports full-body video stitching
5. Supports WebRTC, RTMP, and virtual camera output
6. Supports action choreography: plays custom videos when not speaking
7. Supports multi-concurrency
8. Supports custom digital human avatars
9. Provides frontend API integration

Links

• repo_url: https://github.com/lipku/livetalking
• home_url: https://www.livetalking.ai
• docs_url: https://doc.livetalking.ai