Text models

Google adds Guided Vision to Gemini Live for Android accessibility

Google says Guided Vision in Gemini Live gives spoken descriptions of what your camera sees, built with Aira, for Android 9 and above.

Google’s blog describes Guided Vision, a feature in Gemini Live that uses your camera to give spoken descriptions of your surroundings. Google calls it “a conversational visual partner” and says it was built with the accessibility community.

What to know

  • Google says it can read small print such as nutrition labels, appliance dials and menus in dim light, find objects nearby, and describe colours, patterns and whether clothes match.
  • It can give spoken cues to help you adjust the camera, such as panning, tilting or stepping back.
  • Google says it worked with Aira, which contributed visual interpretation data, and that more than 1,000 members of Aira’s Trusted Tester network tested it.
  • To use it, turn on “Use Guided Vision in Live” in your Gemini profile settings and share your camera in Gemini Live. Google also lists shortcuts in Android Accessibility settings and in TalkBack.
  • Google says it is available on Android 9 and above, in regions and languages where Gemini Live is supported. The post does not mention iOS.
  • Google warns it “can make mistakes” and is “not a medical device, mobility aid, or white cane replacement”. It is not meant for navigation or obstacle detection.

Why it matters

Camera-based assistants are becoming a normal way to prompt: you talk, point and ask follow-ups. Our Gemini Live camera prompts guide has everyday questions to try, and the Gemini tool page tracks the wider model family.

Source: Google blog. We summarise in our own words; check the source for the full details.

Related

More AI news

stay in the loop

New guides, prompts and AI model news by email