Meta's Muse Voice Transcribe Goes Live in LiveKit Agents with 20+ Speaker Diarization

armand_ruiz · x · 2026-09-15

Meta's Muse Voice Transcribe streaming STT model is now available as a plugin in LiveKit Agents. It supports real-time diarization for 20+ speakers, streams mono PCM16 audio at 24 kHz over WebSocket with server-side endpointing, and ships Python/Node.js SDKs — developers only need a Meta Model API key to plug it into an AgentSession, with keyword and language-bias parameters supported.

Original post →

More from coding & agent

coding & agent channel →