On October 1, 2026, Google unveiled Guided Vision within its Gemini Live platform, a groundbreaking feature designed to provide real-time, voice-driven visual assistance. This innovative tool is tailored for individuals who are blind or have low vision, as well as those seeking intuitive visual support. Currently, Guided Vision is accessible on Android devices operating on Android 9 and above, in regions and languages where Gemini Live is available.
The feature was first introduced during the September 1, 2026, Android Drop, where it was highlighted as an upcoming addition for compatible devices. The initial preview detailed the functionality of camera-sharing descriptions and voice feedback for reframing, while also noting that the feature is not infallible and should not be regarded as a medical device or a substitute for safe travel guidance.
How Guided Vision Works
During a session with Gemini Live, users can share their phone’s camera feed, allowing Gemini to provide spoken descriptions of the environment. The system is designed to respond to follow-up questions and offer proactive verbal cues to help users focus on their desired exploration. If the camera is misaligned—whether too high, too close, or slightly off-center—Gemini gently prompts the user to adjust their position, ensuring that it captures the necessary visual context to deliver accurate assistance.
Isha Sheth, Senior Product Manager for Gemini Live, describes Guided Vision as a conversational visual partner that seamlessly integrates into daily routines, enhancing user experience with minimal disruption.
Developed With Aira and the Accessibility Community
Google emphasized that Guided Vision was developed in close collaboration with the accessibility community. The company partnered with Aira to interpret visual data, dedicating tens of thousands of hours to training Gemini Live for real-world conversational contexts. Over 1,000 members of Aira’s Trusted Tester network contributed to refining the model through rigorous testing in everyday scenarios, while Aira specialists collaborated with Google teams to establish safety protocols.
Sally Khoo, Deputy Director (Innovation) at SG Enable, remarked on the significance of such technology, stating, “As standard phones are typically built for sighted users who can see the screen, a product like Gemini Live’s real-time visual descriptions with proactive verbal camera guidance can help bridge a critical gap in everyday tasks at home, from reading fine print to locating daily essentials.”
Google’s commitment to inclusivity is reflected in the extensive feedback gathered from blind and low-vision communities across various countries, including India, Brazil, Singapore, Indonesia, and Japan. This input has been instrumental in ensuring that the descriptions, reframing cues, and responses are both natural and beneficial.
Prashant Verma from the National Association of the Blind in India emphasized the importance of localized testing, stating, “Testing these capabilities directly with blind and low-vision users across Indian languages ensures this technology delivers dependable, practical assistance in everyday life.”
Documented Everyday Use Cases
Google has outlined several everyday scenarios where quick visual verification is essential. These include:
- Reading fine print and complex text, such as nutrition labels, appliance dials, and menus in dimly lit restaurants.
- Locating objects, like a dropped earbud or spices in a crowded cabinet.
- Describing colors, patterns, and shapes, enabling users to determine if clothing items match.
- Exploring unfamiliar environments, providing descriptive overviews of room layouts or objects on a table.
Guided Vision also supports conversations in multiple languages, making it a valuable tool not only for those with visual impairments but also for older adults, individuals with low literacy, or anyone needing assistance with fine print in low-light conditions.
Access Paths, Availability and Stated Limitations
Users can access Guided Vision through the Gemini app, an Android Accessibility shortcut, or Google TalkBack. To enable the feature, users need to navigate to the Gemini app settings, activate “Use Guided Vision in Live,” and then share their camera. Additionally, users can configure Accessibility shortcuts to launch Guided Vision via various gestures or button presses.
It is important to note that while Guided Vision serves as an assistive utility, it is not without limitations. Google advises that the feature can make mistakes and is not intended to replace medical devices, mobility aids, or safe travel practices. Users are encouraged to continue relying on established mobility aids and safe-travel methods in physical environments.
To get started with Guided Vision, Google recommends updating both the Gemini app and the Android system software to their latest versions.
