All models
Microsoft
MAI-Voice-2
Expressive speech for long texts
This model is not in the catalog right now.
See available modelsAbout the model
MAI-Voice-2 is built for long forms: narration, educational materials, voice-overs where natural intonation matters more than output speed. It can handle several speakers within one text and control delivery; according to the publisher, it supports 15 languages in 18 regional variants.
Strengths
- Natural intonation on long texts
- Several speakers within one text
- According to the publisher: 15 languages, 18 regional variants
- Billed per character
Where to use it
- Narration and audiobooks
- Educational materials
- Dialogue voice-overs