AI Matters
Why it matters
Meta releases real-time speech transcription model
The Decoder Meta's new real-time audio model is the foundation for AI assistants that never stop listening
Meta's Superintelligence Labs released Muse Voice Transcribe, an open model that processes speech in 80-millisecond chunks for highly accurate, low-latency transcription.
Why it matters
This technology enables faster, more natural voice assistants that can transcribe speech instantly, distinguish between speakers, and detect sentence ends.
Latest brief · Oct 19, 2026