Meta's first real-time audio model does transcription, speaker separation and turn-taking at once
Three jobs that normally need three components now sit in one autoregressive model, which is the part that matters more than the word error rate.
Model
Muse Voice Transcribe
Muse Voice Transcribe is an artificial intelligence model developed by Meta AI. This page lists what is publicly known about the model and every benchmark result Aivio News has recorded for it, each with its own source and date.
- Developer
- Meta AI
- Model
- Muse
- Released
- September 1, 2026
- Modalities
- Audio, Model
- Status
- Active
- Source
- research.meta.ai
Benchmarks
All storiesNo verified results recorded yet.