What it is: Stimm is an Open Source (AGPL v3) platform designed to orchestrate low-latency voice conversations with AI agents. We focused heavily on the Python/FastAPI event loop and media transport (using LiveKit for WebRTC) to achieve a near real-time feel.

Why we built it: Tired of the 3-second delay in commercial voice bots. Stimm provides a robust foundation for building production-grade telephony and web-based agents.

Key features: Modularity (swap LLMs/STT/TTS easily), built-in SIP support, and Docker integration.

Looking forward to any technical feedback on the architecture!