Google has introduced Guided Vision, an accessibility feature that provides real-time audio descriptions of visual content for users with visual impairments. The tool launches today within Gemini Live on compatible Android devices, allowing users to point their camera at objects, text, or scenes to receive spoken descriptions.

The feature builds on Gemini Live's existing capabilities, extending its functionality beyond conversational AI to include visual interpretation. When activated, Guided Vision analyzes the camera feed and delivers audio feedback about surroundings, text content, or specific objects. Users can also pose follow-up questions about what they're viewing, such as asking for expiration dates on food items or details about identified products.

Google has integrated the functionality across multiple platforms. It operates within the Gemini app interface, appears in Google TalkBack screen reader, and can be triggered through an accessibility shortcut in device settings for Android 9 and newer versions. The company emphasizes that Guided Vision serves as an assistive tool rather than a comprehensive mobility solution, explicitly cautioning against its use for navigation, safety-critical tasks, or as a substitute for mobility aids like canes.

The development reflects Google's broader push to enhance accessibility features across its AI ecosystem. The company joins other tech firms in integrating real-time visual assistance, drawing conceptual parallels to Apple's VoiceOver Live Recognition functionality available on iPhone and Vision Pro devices. Implementation requires no additional hardware beyond a standard Android smartphone with camera access.