MicrosoftMAI Voice
MAI-Voice-2.1
MAI-Voice-2.1 is a public-preview speech API for expressive long-form audio.
Preview
What the release provides
It supports 23 languages, SSML emotion controls and consistent speakers across extended narration. Instant voice cloning is gated and requires a consented reference voice; curated prebuilt voices are also available.MAI-Voice Azure Speech documentation
How to use it well
Practical guidance. Listen to complete sections of the intended narration. Review names, numbers and changes of language with the chosen speaker. Keep the voice identifier and delivery controls with the script configuration so later content edits retain a consistent voice.
Specifications & access
| Detail | Value and conditions |
|---|---|
| Status | Preview |
| Access | API |
| Model identifiers | MAI-Voice-2.1 |