Official provider results
MAI-Transcribe-2-Streaming launch notes
Microsoft launched streaming transcription in 60 languages at an introductory $0.54 per hour through the end of 2026.
Source snapshots refreshed
Five measurement lanesVOICE / S2S
Reasoning, task completion, conversation dynamics, experience, and latency stay separate.
Benchmark profile
- Scope
- Provider-reported languages, availability, introductory audio-duration price, and first-hypothesis timing
- Primary metric
- Language coverage and USD per hour of audio
- Owner
- Microsoft AI
- Available evidence
- Provider specifications published October 1, 2026; evaluator measurements are linked separately
What it measures
Voice experience
Is the exchange natural, robust, and responsive?
Latency
How long do responses, tool calls, and full tasks take?
Available results
- Languages
- 60
- Introductory streaming price
- $0.54 per hour
Microsoft-reported
Continuous automatic language detection.
Through December 31, 2026
Equivalent to $9 per 1,000 minutes. The post-introduction price is not published here.
Interpretation limit
Microsoft reports first hypotheses just over 100 ms after receiving audio. That timer differs from transcript latency measured after detected speech end. The non-streaming model's FLEURS result does not establish this streaming release's accuracy.