Microsoft AI Releases MAI-Transcribe-2-Streaming Streaming Speech-to-Text Model, Tops the Artificial Analysis Real-Time Leaderboard
Microsoft AI Releases MAI-Transcribe-2-Streaming: #1 Real-Time Speech-to-Text Model on Artificial Analysis
AI BriefingOn October 1, 2026, Microsoft AI released its first streaming speech-to-text model, MAI-Transcribe-2-Streaming, along with two speech-generation models: MAI-Voice-2.1 and MAI-Voice-2.1-Flash.
Reported by 5 additional sources
- Microsoft AI Releases the MAI-Transcribe-2-Streaming Real-Time Transcription Model and MAI-Voice-2.1 Speech ModelThe Decoder: AI News · October 2
- Microsoft Releases Its First Real-Time Streaming Speech Transcription Model, MAI-Transcribe-2-Streaming, with a 2.50% Word Error Rate and 0.13-Second Latency, Ranking First on Artificial AnalysisIT Home · Artificial Intelligence · October 2
- Microsoft AI releases MAI-Transcribe-2-Streaming and the MAI-Voice-2.1 family of speech modelsMicrosoft AI: Official Blog · October 2
- Microsoft AI Releases the MAI-Transcribe-2-Streaming Streaming Speech Transcription ModelMustafa Suleyman · October 2
- Microsoft AI Releases the MAI-Transcribe-2-Streaming Streaming Speech Transcription Model, Topping AA-WER Streaming AccuracyArtificial Analysis · October 2
Source: MarkTechPost · marktechpost.com