GoogleFeature UpdatesOfficial Blog
Google Enhances Gemini 2.5 Audio Model for Live Voice Agents
Creating seamless conversational voice AI now requires less manual adjustment.
Key Points
- 1Improved function calling for voice agents
- 2Better adherence to complex instructions
- 3Available in AI Studio, Vertex AI, and more
Google updated Gemini 2.5 Flash Native Audio to improve performance for live voice agents. Function calls to external APIs, instruction-following, and conversation flow were enhanced, easing complex voice dialogue creation. The model is accessible via AI Studio and Vertex AI, though availability varies by product and region.