ModelMAI Voice 2.1
MAI-Voice-2.1
A multilingual expressive speech synthesis foundation model in Microsoft's MAI series. It supports speech generation and can be integrated and invoked through third-party gateways such as the Vercel AI Gateway.
Timeline (last 90 days)
Oct 13
The first batch of MAI models integrated into Vercel AI Gateway includes MAI-Voice-2.1, MAI-Voice-2.1-Flash, MAI-Transcribe-2-Streaming, MAI-Transcribe-2, MAI-Image-2.6, and related variants.
Unverified50%
Oct 13
MAI-Voice-2.1 is priced at $22 per million characters.
Unverified50%