AI Research Atlas

MAI-Transcribe-2-Streaming, MAI-Voice-2.1 and Voice-2.1-Flash

Microsoft · 1 October 2026

Streaming transcription ranks #1 on Artificial Analysis with first hypothesis in ~100 ms; Voice-2.1-Flash has 150 ms end-to-end latency.

Real-time versions for voice agents. Transcribe-2-Streaming is $0.54/hour introductory, Voice-2.1 is $22 per 1M characters (23 languages), and Flash is $15.

Date
Thursday, 1 October 2026
Lab
Microsoft
Kind
model
Access
closed API
Price
Transcribe-2-Streaming $0.54/h (intro); Voice-2.1 $22 and Flash $15 per 1M chars

Figures

MeasureValueMeasured by
Streaming transcript latencyfirst hypothesis ~100 ms
ranked #1 for final and partial transcripts on Artificial Analysis (company-reported)
company

Sources

  1. microsoft.ai/news/our-first-streaming-transcription-model/

This record was checked against its sources on 6 October 2026. How we check

Related

Read the daily brief for 1 October 2026