Muse Voice Transcribe
Muse Voice Transcribe is Meta’s real-time speech-to-text model, released on September 1, 2026.
- Lab
- Meta
- Release date
Meta Superintelligence Labs introduced Muse Voice Transcribe on September 1, 2026 as its first real-time audio perception model, part of the Muse Spark family. It takes streaming audio and returns text, with speaker tags (diarization) and speech onset and endpoint markers. Meta said it was trained on more than 70 languages, 25 of them extensively verified and recommended for the initial release, and supports audio longer than an hour with 20 or more speakers. It is available via the Meta Model API, Meta AI for Mac and Muse Code. Meta did not list an API model ID or a preview or general-availability label.
- Modalities
- Streaming audio input; text output (speech to text, diarization)
- Availability
- Meta Model API, Meta AI for Mac and Muse Code.
- Official source
- research.meta.ai
- Last checked
Questions
When was Muse Voice Transcribe released?
What does Muse Voice Transcribe do?
Where is it available?
Source
Release dates, model IDs and limits come from each lab’s own announcement, changelog, documentation or deprecations page, linked on every model with the date it was last checked. How we pick and check sources: editorial standards.