A screen reader that can take a follow-up question.
Google announced a rollout that let Android TalkBack users ask Gemini follow-up questions about images and about the whole screen, extending beyond one-shot image descriptions.
- Maturity
- Deployed, stage 4 of 4
- Support
- 1 source · institution
- Evidence detail
- How we know ↓
Images without alt text and visually structured app screens can block blind and low-vision users from getting the details they need.
A basic image description can name an object but may not answer the user's own question about material, color, price, or surrounding context.
Gemini generated descriptions and answered user follow-up questions about images and screen contents inside Android's TalkBack screen reader.
What was shown, and what wasn’t
Shown
Google stated that the expanded Gemini integration was rolling out on May 15, 2025, allowing descriptions and questions about images and the entire screen. The same announcement said a new English Expressive Captions version was rolling out for Android 15+ in the United States, United Kingdom, Canada, and Australia.
Not shown · limits
- This is a product announcement, not an independent accuracy or user-outcome study.
- Availability depends on device, operating-system version, language, and region.
- Generated visual descriptions and answers can be incorrect or unsuitable for sensitive content.
Bright editorial interpretation
The interface gives the person asking the question more control over what information is relevant in the moment.
Still open
How accurate and useful are answers in high-stakes tasks such as shopping, travel, or reading personal documents?
How do users want privacy and data handling explained when asking about their screens?
Which languages and device classes remain excluded?
How we know1 source · checked 2026-09-07 · no corrections
Original sources
Institutions: Google · Project Euphonia · University College London Centre for Digital Language Inclusion
- Maturity
- Deployed
- Source published
- 2025-05-15
- Captured
- 2026-09-07
- Last source review
- 2026-09-07
- Editorial method
- AI-assisted source review
Bright compared this account with the linked original and supporting sources and kept reported, budgeted, projected, and observed claims distinct. Bright did not independently audit the underlying records.
Maturity describes the tested or operational setting. Confidence describes support for the particular claim; one does not determine the other.
Revision & correction history
2026-09-07 · The interface gives the person asking the question more control over what information is relevant in the moment.
No corrections recorded.