Model comparison
Compare MAI-Transcribe-2 and Scribe v2 using the same provider-sourced speech recognition & transcription rubric. No mystery score and no invented benchmark ranking.
Facts checked September 4, 2026
Set your usage. Your estimate updates as you type.
Uses the recording length in hours: 30 minutes = 0.5 hours. Extra features and minimum charges may change the bill.
Estimates exclude taxes, tools, cache storage/writes, free allowances and custom discounts. Image estimates cover output only, not prompt or reference-image charges. Quality modes differ by model. Unlisted settings are not treated as free.
| Model | Access | Estimated total (USD) |
|---|---|---|
| MAI-Transcribe-2Microsoft | Microsoft | No reviewed rate |
| Scribe v2ElevenLabs | ElevenLabs | No reviewed rate |
Quick take
Microsoft's fast batch transcription model for long recordings, speaker labels, word timing, keyword biasing, and clean or verbatim transcripts.
Microsoft's one-hour-in-about-ten-seconds figure is a provider claim, not a guaranteed SLA. Benchmark your own audio and confirm the post-preview price, deployment region, quotas, and retention configuration.
ElevenLabs' multilingual speech-to-text model for detailed transcripts with many speakers, word timing, audio events, and large custom keyterm lists.
Do not assume batch and realtime are the same route or price. Zero-data-retention is conditional rather than automatic, so verify the plan, endpoint, region, and enable_logging behavior before processing sensitive audio.
Compare the published facts
Values use each provider's own published units and limits. A blank means the provider did not publish a directly comparable value in the sources reviewed.
| Speech recognition & transcription | MAI-Transcribe-2 | Scribe v2 |
|---|---|---|
| Transcription priceCurrent provider price per audio hour or minute for the listed processing route. | $0.10 / audio hour limited-time rate | $0.22 / audio hour; realtime listed separately |
| Live or batchWhether the model handles realtime streams, uploaded recordings, or both. | Fast batch and long-form transcription | Batch and long audio; separate realtime route |
| LanguagesProvider-published language coverage, separating trained or advertised coverage from specifically verified languages where needed. | 60 languages | 90+ languages |
| Speaker labelsWhether the model identifies who spoke and any published speaker limit. | Yes | Up to 32 speakers |
| TimestampsAvailable word-, segment-, or utterance-level timing information. | Word-level timestamps | Word-level timestamps |
| Vocabulary controlKeyword boosting, custom spelling, context, prompting, or other ways to improve domain terms. | Keyword biasing, clean/verbatim output, automatic language detection | Up to 1,000 keyterms; audio tags and language detection |
| Where to use itDirect API, cloud catalog, application, or regional endpoint documented by the provider. | Microsoft Foundry / Azure | ElevenLabs Speech-to-Text API |
How to choose
Start with the job you need to complete, then validate cost, access, and policy details on your exact provider route.
Microsoft says prompts, outputs, embeddings, and training data submitted to Foundry Models are not available to model providers and are not used to train foundation models without permission. Retention and abuse-monitoring details depend on the deployed service.
ElevenLabs lets eligible API customers manage data-use preferences. Zero-data-retention through enable_logging=false is limited to qualifying enterprise configurations and supported models, including Scribe v2; verify the live plan and regional endpoint.
Provider and API links
Frequently asked questions
MAI-Transcribe-2: Microsoft's fast batch transcription model for long recordings, speaker labels, word timing, keyword biasing, and clean or verbatim transcripts. Scribe v2: ElevenLabs' multilingual speech-to-text model for detailed transcripts with many speakers, word timing, audio events, and large custom keyterm lists.
Consider MAI-Transcribe-2 when your priority is High-volume recorded calls and meetings. Consider Scribe v2 when your priority is Multilingual media transcription. Test both with your own data and provider route before committing.
No. This comparison aligns provider-published facts for the Speech recognition & transcription category. It does not claim a universal winner or combine incompatible third-party benchmark scores.
