MicrosoftAudio matchupSpeech to text
MAI-Transcribe-2vs.MAI-Transcribe-1.5
Which model is better?
MAI-Transcribe-2 is slightly ahead in independent tests, but the lead is not certain.
Independent testsArtificial Analysis
MAI-Transcribe-20 of 1
MAI-Transcribe-1.50 of 1
Popularityout of 100
MAI-Transcribe-20
MAI-Transcribe-1.50
Weights and access
MAI-Transcribe-2proprietary
MAI-Transcribe-1.5proprietary
Priceverbatim from the provider
MAI-Transcribe-2$0.10 per hour of audio as an introductory price
MAI-Transcribe-1.5$0.36 per hour of audio
Pick MAI-Transcribe-2 for
leaning Artificial Analysis
Pick MAI-Transcribe-1.5 for
No clear edge
Shared tests
Independent evaluators and leaderboards only, the newest shared snapshot per evaluator. With 95 % ranges a lead counts when they do not overlap; for hit rates, when it holds in a 95 % test. Without any uncertainty figure at most “slightly ahead”.
Artificial Analysis
MAI-Transcribe-2 slightly aheadArtificial Analysis AA-WER v2.0 (non-streaming)
MAI-Transcribe-22.0 %
MAI-Transcribe-1.52.4 %
lower is better, as of October 16, 2026
Source · latest October 16, 2026
All data side by side
Prices are verbatim from the provider, because audio models bill per minute, character or token.
| MAI-Transcribe-2 | MAI-Transcribe-1.5 | |
|---|---|---|
| Task | Speech to text | Speech to text |
| Developer | Microsoft | Microsoft |
| Status | Preview | Preview |
| Released | September 3, 2026 | June 2, 2026 |
| Weights | proprietary | proprietary |
| Runs locally | no | no |
| Price | $0.10 per hour of audio as an introductory price | $0.36 per hour of audio |
| Popularity (out of 100) | 0 | 0 |