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 / A self-hosted AI streamer that sees and hears the screen
Wallie V2
An MIT-licensed AI streamer combining screen and audio context, configurable personas, speech output, Twitch/YouTube/Kick chat and a VTube Studio avatar connection.
View repository ↗Source code</>
Wallie V2
The unlock
Why this one is on the map.
It exposes how perception, memory, voice and audience messages can be joined into a continuing character-led stream.
Screen context and chat give a self-hosted character live material.
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.
Inspect the repository, configure the chosen backends and connect the avatar and broadcast software separately.
Best for: Builders of autonomous hosts and reactive AI commentators.
Access note: Self-hosted code; backend, avatar and input permissions require setup.
Sources & status
Keep the link, keep the caveat.
2026-10-29: Official README and MIT license reviewed. Requires local setup, selected speech/model backends and separate avatar or broadcast configuration. No code was executed, keys supplied or autonomous stream tested.
Explore similar projects
Suggest a correction ↗Source code</>
WorldCupVoice
Experimental
WorldCupVoice
An open-source project that watches an Agora RTC live feed, generates context-aware voice commentary, and publishes the AI audio back to viewers with a transcript.
Source code</>
UNREEL
Experimental
UNREEL
An open-source, self-hosted AI streaming service that writes scenes and renders MiniMax H3 Max Turbo video through fal as a channel plays.