Meta Superintelligence Labs Launches Muse Voice Transcribe, First Real-Time Audio Perception Model
aigclink · x · 2026-09-02
Meta Superintelligence Labs has introduced Muse Voice Transcribe, described as its first real-time audio perception model. Key capabilities:
- Real-time streaming ASR with endpointing;
- Diarization supporting 20+ speakers;
- 25+ languages with seamless code-switching;
- Improved accuracy via language, keyword and context biasing.
Meta says the model ranks first on Artificial Analysis streaming speech-to-text and on public diarization benchmarks (as of September 1, 2026). A live microphone transcription demo is available; audio is not stored.
More from Models
- Mocked benchmark shows DeepSeek's document chart grounding badly misaligned — burkov · 2026-10-18
- Dev complains Claude's habit of saying "not done" is his biggest pain point — jasonkneen · 2026-10-18
- Open-weight TYGER speech decision model makes 80ms turn-taking calls for full-duplex voice agents — solyarisoftware · 2026-10-18
- Paying Google AI subscriber blocked: custom-domain Workspace accounts can't access AI Studio — ChrisRackauckas · 2026-10-18
- Running Ling 3.1 Flash (560B MoE, 25B Active) Across 4 DGX Sparks via llama.cpp RPC — BerryBlush_Rogers691 · 2026-10-18
- Users press OpenAI on long-delayed Adult Mode a year after it was promised — ___Patrice___ · 2026-10-18