Source code</>
LiveTalking
Experimental
LiveTalking
A digital-human conversation engine supporting several lip-sync backends, interruption handling and WebRTC, RTMP or virtual-camera output.
Open source / Agents that understand live video
Vision Agents by Stream
A framework for building agents that watch and listen to live video, combining perception models with real-time conversation.
View repository ↗Source code</>
Vision Agents by Stream
The unlock
Why this one is on the map.
Adjacent perception layer for an AI co-host or coach that reacts to what is on screen; not a video generator.
The clearest product signal is “Let the agent see the action”. That interaction model—not a generic AI feature list—is the reason to compare it with the rest of this directory.
This is source code, not a hosted channel. You need to inspect, configure, and deploy it with your own keys and runtime before anything can broadcast.
How to read it
What happens when you open it.
Open the official destination and look for the current adjacent infrastructure · live perception · agent sdk workflow. Controls, access and live activity can change after this editorial check.
Best for: Agents that understand live video
Access note: Read the repository setup and license before running. Hardware, model downloads and API access vary by project.
Sources & status
Keep the link, keep the caveat.
2026-09-05: Official README reviewed. Video provider and model configuration required; no measured latency or autonomous broadcasting claim.
Explore similar projects
Suggest a correction ↗Source code</>
Open-LLM-VTuber
Experimental
Open-LLM-VTuber
A voice-interactive AI companion with visual perception and a Live2D character, supporting local and hosted model backends.
Source code</>
MuseTalk
Experimental
MuseTalk
A lip-sync model that updates a prepared face video from incoming audio clips, with a documented real-time inference workflow.