Official provider results
MAI-Transcribe-2 launch evaluation
Microsoft AI’s September 3, 2026 launch post reports FLEURS word error rate across 60 languages, an independent leaderboard position, and relative speed against competing transcription models.
Source snapshots refreshed
Five measurement lanesVOICE / S2S
Reasoning, task completion, conversation dynamics, experience, and latency stay separate.
Benchmark profile
- Scope
- FLEURS average WER over 60 languages, independent WER leaderboard rank, and throughput comparisons versus GPT-Transcribe, Scribe v2, and Gemini 3.5 Transcribe
- Primary metric
- Word error rate (lower is better)
- Owner
- Microsoft AI
- Available evidence
- Exact figures published in the official launch post
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
- FLEURS average WER (60 languages)
- 5.2%
- Independent WER leaderboard rank
- #2
- Speed vs GPT-Transcribe
- 10x faster
- Price
- $0.10 per hour
Ranked first by Microsoft
Lower is better.
Provider-reported
Microsoft also claims the accuracy–latency Pareto frontier.
Provider-reported
Also 7x faster than ElevenLabs Scribe v2 and 5x faster than Gemini 3.5 Transcribe.
Limited-time rate
Down from $0.36 per hour for MAI-Transcribe-1.5.
Interpretation limit
These are provider-reported launch results; the post gives competitor rankings and speed ratios without their exact WER values.