Google launches Guided Vision in Gemini Live to describe your surroundings in real-time

Google has rolled out Guided Vision, a real-time audio description feature now available in Gemini Live on Android devices. Users can point their phone camera at objects, text, or environments and receive spoken descriptions and assistance from the AI.

How Guided Vision Works#

By activating camera sharing in Gemini Live, users can ask Gemini to help with a range of tasks: deciphering small print, describing surroundings, locating or naming objects, and providing details about specific items. The feature includes audio cues to help users position their camera correctly if the target object isn't initially in view, and users can pose follow-up questions—such as asking Gemini to read an expiration date on a package it has identified.

Accessibility Integration#

Guided Vision is built into the Gemini app itself, but Google also integrated it into TalkBack, the Android accessibility tool, and made it available as a shortcut in the Settings app on devices running Android 9 and later. This follows Google's earlier accessibility announcements during its recent Pixel product lineup updates.

The tool joins comparable offerings like Apple's VoiceOver Live Recognition feature, which provides similar real-time visual assistance on newer iPhones and Vision Pro.

Important Limitations#

Google emphasizes that users should not rely on Guided Vision for critical safety tasks. The company specifically cautions against using the tool for navigation, safe-travel guidance, obstacle detection, or as a substitute for mobility aids like white canes. The feature is intended to supplement rather than replace specialized assistive technology.