Google launches Guided Vision in Gemini Live for real-time spoken visual assistance

Google launched Guided Vision in Gemini Live on October 1, 2026, adding real-time visual interpretation to voice conversations. Users can share their camera during a Live session to hear descriptions of their surroundings, ask follow-up questions and get spoken prompts—such as moving the camera right or tilting it down—to provide clearer visual context. Google says the feature is intended for people who are blind or have low vision, as well as others who want visual assistance.

Google says Guided Vision was developed alongside the accessibility community. It partnered with Aira and used tens of thousands of hours of visual data to train Gemini Live for conversational, real-world context. More than 1,000 members of Aira’s Trusted Tester network helped test and refine the model, while Aira specialists advised on safety guardrails. Google also says it had previewed the feature earlier.

Examples Google lists include reading fine print, finding nearby objects, identifying colors or patterns, and getting descriptions of unfamiliar spaces. The feature supports multiple languages, with Google citing testing and feedback from blind and low-vision communities in India, Brazil, Singapore, Indonesia, Japan and elsewhere. Users can enable it in Gemini app settings, or access it through Android Accessibility Shortcuts and TalkBack. It is available on Android 9 and later in regions and languages where Gemini Live is supported.

Google cautions that Guided Vision can make mistakes. It is not a medical device, mobility aid or white-cane replacement, and it is not intended for navigation, safe-travel guidance or obstacle detection. That distinction defines the feature’s role: it can help users interpret what a camera sees, but it is not a substitute for established mobility aids or safe-travel practices.