Meta's new real-time audio model is the foundation for AI assistants that never stop listening

Meta's Superintelligence Labs has launched Muse Voice Transcribe, a real-time transcription model that processes speech in 80-millisecond chunks. The model distinguishes between speakers and detects sentence boundaries, enhancing its utility in diverse applications. According to Artificial Analysis, Muse Voice Transcribe delivers the most accurate streaming transcription at the lowest price in the market. Meta envisions this model as a foundational component for personal AI agents that can listen in on real conversations through devices like its camera glasses.