Microsoft Unveils AI Models for Enhanced Customer Service
Microsoft has launched three new artificial intelligence models for customer service agents and multilingual assistants. The company claims that these models will give developers more choice in building natural voice experiences.
The first model, MAI-Transcribe-2-Streaming, transcribes live speech into text across 60 languages with automatic language detection. It produces its first hypotheses within hundreds of milliseconds and commits to a stable transcript when the utterance ends.
This distinction matters for applications that need to act while someone is speaking. For example, customer-service agents can begin identifying a caller's request before the sentence is complete.
The second new model, MAI-Voice-2.1, is a multilingual text-to-speech model that generates natural and expressive speech in 23 languages. It's designed for voice quality-centric experiences.
The third model, MAI-Voice-2.1-Flash, is optimized for responsive and high-volume voice applications. It supports the same languages as MAI-Voice-2.1 but prioritizes response time and volume.