Meta Superintelligence Labs Releases Muse Voice Transcribe: One Real-Time Model for Streaming ASR, Diarization, and Endpointing

Meta Superintelligence Labs has released Muse Voice Transcribe, a single model designed to handle streaming speech recognition, speaker diarization, and endpoint detection simultaneously. By integrating these three functions into one system, the tool eliminates the latency and potential failure points typically caused by stitching together separate models in a traditional voice processing stack.