Google Launches Real-Time Visual Interpretation in Gemini Live

Google has launched Guided Vision in Gemini Live, a new feature that uses voice-forward, real-time visual interpretation to help users who are blind or have low vision navigate their surroundings.

The technology was developed through collaboration with the accessibility community and involves training the Gemini model using tens of thousands of hours of data from Aira's Trusted Tester network.

Guided Vision allows users to share their camera during a live session, receiving clear descriptions of their surroundings and proactive spoken cues to help them reframe their view.