As of: 2026-11-19 22:11 CET. Researched and written by AI agents (Claude Opus 5.5 in Claude Code), with a human editor. How we work: https://postcutoff.com/about/ Canonical page: https://postcutoff.com/m/mai-voice-2-1/ # MAI-Voice-2.1 / MAI-Voice-2.1-Flash Microsoft, MAI-Voice family. Status: current. Type: speech and audio. Released: 2026-10-01. ## How to call it | Provider | Model id | Endpoint | Docs | |---|---|---|---| | Microsoft Foundry | | | https://microsoft.ai/news/our-first-streaming-transcription-model/ | | OpenRouter | | | https://openrouter.ai/ | | Web app (MAI Playground) | | | https://playground.microsoft.ai/ | ## Pricing - Per 1M characters: $22 - Unit as published: USD per 1M characters (MAI-Voice-2.1); MAI-Voice-2.1-Flash is $15 per 1M characters - Source: https://microsoft.ai/news/our-first-streaming-transcription-model/ - Last checked: 2026-10-02 ## Specs - Input: text - Output: audio - Context window: not published - Max output: not published - Open weights: no - Licence: proprietary ## Where it stands - Newer speech and audio models in the registry since its release: 11. The 5 newest: - NVIDIA Nemotron 3.5 Transcribe (30B-A4B) (2026-10-28) https://postcutoff.com/m/nemotron-3-5-transcribe/ - Hy-SimulMT-30B-A3B (2026-10-22) https://postcutoff.com/m/hy-simulmt-30b-a3b/ - Eleven Realtime v1 Max / Mini (2026-10-20) https://postcutoff.com/m/eleven-realtime-v1-max/ - Voxtral Mini Transcribe 3 (+ Mini Transcribe Realtime 3) (2026-10-20) https://postcutoff.com/m/voxtral-mini-transcribe-3/ - Voxtral Small Transcribe 3 (2026-10-20) https://postcutoff.com/m/voxtral-small-transcribe-3/ - Before it from Microsoft: MAI-Transcribe-2 (2026-09-03) https://postcutoff.com/m/mai-transcribe-2/ - Closed weights. ## What stands out - Cross-lingual voice with native accent: 23 languages / 26 locales; a single voice keeps its native accent across languages; consent-gated voice cloning. Source: https://microsoft.ai/news/our-first-streaming-transcription-model/ - Low-latency Flash variant: MAI-Voice-2.1-Flash: 150 ms end-to-end latency, 55% faster inference, $15 per 1M characters. Source: https://microsoft.ai/news/our-first-streaming-transcription-model/ ## Notes Successor to MAI-Voice-2 (Build, June 2026). Exact API model ids not verified. Source: https://microsoft.ai/news/our-first-streaming-transcription-model/ Model page: https://postcutoff.com/m/mai-voice-2-1/