LiveKit showcases support call agent using Microsoft AI speech models and Gemma 4

LiveKit demonstrated a support call workflow built with LiveKit Agents and Microsoft AI speech models.

In the pipeline, MAI-Transcribe-2-Streaming handles real-time audio input ([STT](https://livek.it/E4qoZvU)), Gemma 4 deployed on LiveKit Inference manages reasoning and tool calls, and MAI-Voice-2.1-Flash provides spoken output ([TTS](https://livek.it/RsCc37l)).