Microsoft launches MAI voice lineup; streaming transcription tops AA leaderboard

On October 2, Microsoft AI chief Mustafa Suleyman announced MAI-Transcribe-2-Streaming, a streaming speech transcription model, alongside MAI-Voice-2.1 and MAI-Voice-2.1-Flash, two voice models targeting more natural speech and shorter conversational turn latency for voice agents. The three models have since gone live on LiveKit.

Confirmed

Why it matters

2026-10-02 ~ 2026-10-03 · 6 related posts

Primary sources