Meta Superintelligence Labs Unveils Muse Voice Transcribe, a Real-Time Audio Perception Model

AI at Meta ·

Key Info

Meta Superintelligence Labs has launched Muse Voice Transcribe, a real-time audio perception model that performs streaming speech-to-text, diarization for over 20 speakers, and endpointing, with multilingual code-switching and biasing features.

Highlights

  • Real-time streaming ASR with speaker diarization supporting 20+ speakers and automatic endpointing.
  • Multilingual with seamless code-switching; accuracy improves via language, keyword, and context biasing.
  • Ranks first on Artificial Analysis streaming speech-to-text and public diarization benchmarks.
Loading...