Current model · Microsoft AI
MAI-Transcribe-2-Streaming benchmarks, pricing and specs
MAI-Transcribe-2-Streaming is a Microsoft AI model released on October 1, 2026 that takes audio input and produces text. It costs $0.009 per audio minute. It scores 2.51 % WER on Artificial Analysis AA-WER Streaming, #1 of 4 current models (Artificial Analysis, measured November 15, 2026).
- Per audio minute
- $0.009Listed today
- Artificial Analysis AA-WER Streaming
- 2.51 % WER#1 of 4 current models
Data verified November 15, 2026 · Dataset 2026-11-15
Specs
Key facts
- Provider
- Microsoft AI
- Model family
- MAI-Transcribe
- Release date
- October 1, 2026
- Status
- Current
- Access
- API only
- API model id
- MAI-Transcribe-2-Streaming
- Context window
- Not reported
- Max output
- Not reported
- Input
- audio
- Output
- text
- Task families
- audio
- Declared capabilities
- speech to text
Charts
Against every current audio model
Where MAI-Transcribe-2-Streaming lands among the current models of its task, on every independent benchmark, on price and on context.
Where it stands
On Artificial Analysis AA-WER Streaming, MAI-Transcribe-2-Streaming ranks #1 of 4 current audio models. 12 of 16 list a lower price per audio minute. Each bar is the share of current models it beats.
Independent benchmarks
Artificial Analysis AA-WER Streaming
#1of 4
2.51% WER · ahead of 100%
Price and context
Price per audio minute
#13of 16
$0.009 · cheaper than 13%
Price
MAI-Transcribe-2-Streaming pricing
It is the #3 most expensive of 16 current audio models priced per audio minute.
Benchmarks
Independent benchmark results
Scores come from independent evaluators that ran the model themselves, pinned to a dated snapshot. Rank is among current models in the same snapshot. Microsoft AI's own launch figures are listed under strengths with the publisher named, because they are not measured the same way.
| Benchmark | Score | Rank | Measured | Source |
|---|---|---|---|---|
| Artificial Analysis AA-WER Streaming | 2.51 % WER | #1 of 4 | November 15, 2026 | Artificial Analysis · 2026-11-15 |
Fit
Strengths, cautions and best uses
- Strengths
- Microsoft reports first partial transcripts just over 100 ms after audio arrives, and first place for accuracy on final and partial transcripts at Artificial Analysis
- 60 languages with automatic, continuous language detection
- $0.54 per hour of audio at launch
- Cautions
- Microsoft Foundry lists it as a preview
- The $0.54 per hour price is introductory until the end of 2026, and the later price is not published
- Best for
- Live captions and voice agents
- Multilingual calls where speakers switch languages
From benchmark to production
Building on MAI-Transcribe-2-Streaming?
Public scores rarely hold on your own data. We test it against the alternatives on your real tasks, then build the agent or pipeline around the model that wins.
AI Model Cost AuditLaunch price $100, regular price $150
Alternatives
Alternatives to MAI-Transcribe-2-Streaming
The closest current audio models by price. The cheapest model that scores as well or better is listed first; failing that, the cheapest one within 3% of the benchmark's scale.
On score and price
MAI-Transcribe-2-Streaming beside the 4 alternatives listed below, on Artificial Analysis AA-WER Streaming and on price per audio minute. 3 of them cost less and score as well or better.
Artificial Analysis AA-WER Streaming
Measured November 15, 2026, higher is better
- MAI-Transcribe-2-Streaming
- 2.5
- Grok Voice Transcribe 2.0
- 2.7
- GPT-4o Transcribe
- Not scored
- Scribe v2 Realtime
- 3.6
- Scribe v3 Live
- 2.9
Price per audio minute
Listed today, lower is better
- MAI-Transcribe-2-Streaming
- $0.009
- Grok Voice Transcribe 2.0
- $0.00167
- GPT-4o Transcribe
- $0.006
- Scribe v2 Realtime
- $0.00367
- Scribe v3 Live
- $0.00367
Cheaper, similar score
Grok Voice Transcribe 2.0
- 81% cheaper per audio minute, $0.001667 vs $0.009
- 0.2 points lower on Artificial Analysis AA-WER Streaming, 2.73 vs 2.51
- From xAI
GPT-4o Transcribe
- 33% cheaper per audio minute, $0.006 vs $0.009
- From OpenAI
Scribe v2 Realtime
- 59% cheaper per audio minute, $0.003667 vs $0.009
- 1.1 points lower on Artificial Analysis AA-WER Streaming, 3.59 vs 2.51
- From ElevenLabs
Scribe v3 Live
- 59% cheaper per audio minute, $0.003667 vs $0.009
- 0.4 points lower on Artificial Analysis AA-WER Streaming, 2.86 vs 2.51
- From ElevenLabs
Further reading
More on MAI-Transcribe-2-Streaming
Compare MAI-Transcribe-2-Streaming with
- MAI-Transcribe-2-Streaming vs Gemini 3.5 Transcribe
- MAI-Transcribe-2-Streaming vs GPT-4o Transcribe
- MAI-Transcribe-2-Streaming vs GPT Transcribe
- MAI-Transcribe-2-Streaming vs Grok Voice Transcribe 2.0
- MAI-Transcribe-2-Streaming vs Scribe v2 Realtime
- MAI-Transcribe-2-Streaming vs MAI-Transcribe-2
Other MAI-Transcribe models
FAQ
MAI-Transcribe-2-Streaming FAQ
How much does MAI-Transcribe-2-Streaming cost?
MAI-Transcribe-2-Streaming costs $0.009 per audio minute at the standard tier. It is the #3 most expensive of 16 current audio models priced per audio minute. Source: Microsoft AI, verified October 29, 2026.
How does MAI-Transcribe-2-Streaming score on independent benchmarks?
MAI-Transcribe-2-Streaming scores 2.51 % WER on Artificial Analysis AA-WER Streaming, #1 of 4 current models (Artificial Analysis, measured November 15, 2026). These are independent evaluations, not Microsoft AI's own launch figures.
Is MAI-Transcribe-2-Streaming open source or API only?
MAI-Transcribe-2-Streaming is available through Microsoft AI's API only; its weights are not published.
What are the best alternatives to MAI-Transcribe-2-Streaming?
The closest current alternatives by price and measured performance are Grok Voice Transcribe 2.0 (81% cheaper per audio minute, $0.001667 vs $0.009), GPT-4o Transcribe (33% cheaper per audio minute, $0.006 vs $0.009), Scribe v2 Realtime (59% cheaper per audio minute, $0.003667 vs $0.009) and Scribe v3 Live (59% cheaper per audio minute, $0.003667 vs $0.009).
AI Model Brief
The AI models worth knowing, once a week
What shipped, how it scores on independent benchmarks, what it costs per million tokens and what was deprecated, from the data behind our AI model benchmarks.
We use your address only to send the brief. See our privacy policy.
See this week’s briefSources
Sources
- Our first streaming transcription model debuts at no. 1 on Artificial Analysis · Microsoft AI · official · verified October 29, 2026
- MAI-Transcribe-2-Streaming in the Microsoft Foundry model catalog · Microsoft · official · verified October 29, 2026
- Artificial Analysis speech-to-text leaderboard (AA-WER Streaming) · Artificial Analysis · independent · verified November 15, 2026 · Artificial Analysis Terms of Use
All 283 sources behind the benchmarks, with their versions, licenses and notes, are in the source catalog.
Cite this page
Van Data Team. "MAI-Transcribe-2-Streaming Benchmarks, Pricing and Specs." AI Model Benchmark Dataset 2026-11-15. https://vandatateam.com/ai-model-benchmarks/mai-transcribe-2-streaming
Reuse the data: JSON · CSV · or explore every model in the AI model benchmark comparison.
Production evaluation
Deciding whether MAI-Transcribe-2-Streaming fits your workload?
Public benchmarks narrow the shortlist. We test the finalists on your own tasks, data and cost limits, then build the routing and review controls to run the winner in production.
- Evaluation cases and acceptance thresholds tied to real work
- Model routing, fallbacks and cost controls
