LiveKit
An open source WebRTC platform for building realtime voice, video and AI agent applications, with a distributed media server, client SDKs for many platforms, and an agents framework for voice AI.
LiveKit is an open source platform for building voice, video and physical AI agents. This repository is the LiveKit server, a scalable, distributed WebRTC SFU (Selective Forwarding Unit) that carries realtime audio, video and data between people, devices and AI models.
It supports client SDKs for web, mobile, desktop, embedded and server use, JWT authentication, UDP/TCP/TURN networking, and deployment as a single binary, Docker container or Kubernetes cluster. Additional features include speaker detection, simulcast, selective subscription, moderation APIs, end-to-end encryption, SVC codecs, telephony over SIP and webhooks.
A companion LiveKit Agents SDK, available for Python and Node.js, adds speech-to-text, language model, text-to-speech and turn detection components for building realtime voice agents. The server itself is written in Go, using the Pion WebRTC implementation.