Agent & Voice Infrastructure

Low-latency, fully modular voice-agent pipeline integrating voice activity detection, speech-to-text, LLMs, and text-to-speech via an OpenAI Realtime-compatible WebSocket API. Apache 2.0 licensed.

Framework for building realtime voice AI agents with WebRTC transport, STT-LLM-TTS pipelines, and production-grade orchestration. Used by Salesforce Agentforce and Tesla. Apache-2.0 licensed.

Open-source framework for voice and multimodal conversational AI. Build real-time voice agents with support for speech-to-text, LLMs, text-to-speech, and live video. BSD-2-Clause licensed.

Web app for interacting with any LangGraph agent (Python & TypeScript) via a chat interface. Stream messages, handle interruptions, and view agent state. MIT licensed.