Microsoft AI releases new transcription and text-to-speech models for voice agents
Microsoft released MAI-Transcribe-2-Streaming and two MAI-Voice text-to-speech models for low-latency voice agents.
Microsoft AI released MAI-Transcribe-2-Streaming for real-time transcription in 60 languages, saying it ranks first for accuracy on Artificial Analysis and returns first partial results in just over 100 milliseconds. Introductory pricing is $0.54 per audio hour through year-end. It also released MAI-Voice-2.1, which speaks 23 languages in one voice, and MAI-Voice-2.1-Flash, cited at 150 milliseconds latency and $15 per million characters instead of $22. Both voice models clone a voice from a few seconds of audio, include misuse safeguards, and are available through Microsoft Foundry, MAI Playground, and OpenRouter.
- MAI-Transcribe-2-Streaming covers 60 languages with first partials in about 100 ms.
- Microsoft claims top transcription accuracy on Artificial Analysis.
- MAI-Voice-2.1-Flash costs $15 per million characters at 150 ms latency.
- Voice models can clone speech from a few seconds of reference audio.
Full article217 words · extracted from the-decoder.com · click to collapse
Microsoft AI has released MAI-Transcribe-2-Streaming, a new model for real-time transcription. Microsoft says it ranks first for accuracy on Artificial Analysis. The model transcribes 60 languages and delivers its first partial results in just over 100 milliseconds. Microsoft says this lets voice agents respond while someone is still mid-sentence. Through the end of the year, an hour of audio costs $0.54 at the introductory price.
Microsoft also released two new text-to-speech models. MAI-Voice-2.1 is designed to speak 23 languages in the same voice, with a native accent in each one. According to Microsoft, the MAI-Voice-2.1-Flash variant hits a latency of 150 milliseconds and costs $15 per million characters instead of $22.
Both voice models can clone a voice from just a few seconds of reference audio, and built-in safeguards are meant to prevent misuse. The models are available through Microsoft Foundry and the MAI Playground, among other platforms, and the two voice models are also on OpenRouter. In one test, about half of the 4,000 participants thought the voices belonged to a real person.
AI News Without the Hype – Curated by Humans
Subscribe to THE DECODER for ad-free reading, a weekly AI newsletter, our exclusive "AI Radar" frontier report six times a year, full archive access, and access to our comment section.
Text extracted automatically; images, tables and formatting may be missing. Original: https://the-decoder.com/microsoft-ai-releases-new-transcription-and-text-to-speech-models-for-voice-agents/