MAI-Transcribe-2-Streaming
MAI-Transcribe-2-Streaming is a Microsoft transcription service that converts live speech to text in real time, delivering tentative text about 100 ms after speech begins and supporting 60 languages. Recent coverage notes its inclusion in a new suite of voice AI tools announced on October 1 2026 and its launch alongside two other voice models.
MAI-Transcribe-2-Streaming is an audio model from Microsoft, first covered on 3 October 2026. It is available through an API.
Last updated
At a glance
- Made by
- Microsoft
- Type
- Audio model
- Availability
- API
- Licence
- Not stated
- Context window
- Not stated
- First seen
- 3 October 2026
Taken from the coverage, not from our own testing; "not stated" means none of the reports we read gave it. All tracked models →
4 stories about MAI-Transcribe-2-Streaming
-
Anthropic invests $100 million to train 10,000 engineers
anthropic announced the "claude frontier academy" and plans to spend $100 million training 10,000 engineers to build talent capable of deploying advanced ai systems. google’s latest frontier model, "gemini 4 argon",…
1 sourcenote.com -
AI that turns into text before you finish speaking: What changes with Microsoft's new voice AI?
Microsoft announced on October 1, 2026 a new suite of voice AI tools that can return text while a speaker is still talking. The transcription service, called MAI-Transcribe-2-Streaming, shows tentative text about 100…
1 sourcenote.com -
GitHub Copilot public preview adds desktop screen control on macOS and Windows
GitHub released a public preview of Computer Use for Copilot CLI and the Copilot app on October 1. The feature lets the assistant read screen content and perform clicks, typing, scrolling, and drag operations across…
1 sourcenote.com -
Microsoft launches three real-time voice AI models for developers
Microsoft released three new AI voice models on Thursday. MAI-Transcribe-2-Streaming converts live speech to text in 60 languages within milliseconds, enabling real-time responses. MAI-Voice-2.1 generates expressive…
10 sourcesMarkTechPostnote.comtechtimes.comThe Decoder +5 more
