MAI-Voice-2.1
Highest-fidelity expressive text-to-speech model across 23 languages, used via Azure Speech SSML.
Specs
- Context
- —
- Max output
- —
- Input price
- —
- Output price
- —
- Released
- —
Context
- Not published
Pricing
- Not published
Modalities
- Input
- Text
- Output
- Audio
API
- API types
Reasoning
- Not published
Info
- Status
- Preview
- Released
- —
- Knowledge cutoff
- —
Features
- Not published
How to call
1 provider
microsoftmodel = MAI-Voice-2.1
audio
POSThttps://<region>.tts.speech.microsoft.com/cognitiveservices/v1
JSON
Standard format
{
"id": "mai-voice-2.1",
"object": "model",
"created": null,
"owned_by": "microsoft",
"name": "MAI-Voice-2.1",
"api": {
"types": [
"audio"
]
},
"limits": {
"context": null,
"input": null,
"output": null
},
"modalities": {
"input": [
"text"
],
"output": [
"audio"
]
},
"reasoning": {
"supported": null,
"efforts": []
},
"pricing": null,
"features": [],
"info": {
"status": "preview",
"release_date": null,
"knowledge_cutoff": null,
"description": "Highest-fidelity expressive text-to-speech model across 23 languages, used via Azure Speech SSML.",
"docs": "https://learn.microsoft.com/en-us/azure/ai-services/speech-service/mai-voices",
"verified_at": "2026-10-11"
}
}Official sources
2