Chrome To Use AI To Aid Blind, Low-Vision Users

Chrome is set to make browsing and navigating through the internet easier for the visually impaired and those who have low vision. Most of the content available in the virtual space is composed of visuals, which is difficult for the blind to understand. Google is addressing this issue by using AI and machine learning for image recognition on Chrome.

Chrome Helps Blind

Chrome will use AI to recognize the images, which will provide textual descriptions of what the image depicts. Sometimes, the content creators use alt text to describe the images used. However, it is still difficult for most blind people using Braille displays or screen readers. Often, they're unable to understand what the creator is trying to show.

An unfortunate scenario is that there are millions of images that are still unlabelled across the internet. Laura Allen, a senior program manager on the Chrome accessibility team, further explains the sorry state for the visually impaired. When they're navigating via a screen reader of Braille display, often the user will simply hear 'image' or 'unlabelled graphic'. Allen also has low vision and she hears something like "super long string of numbers which is the file name, which is just totally irrelevant," she says.

Chrome New Feature Uses AI

The new Chrome feature is based on the same technology that allows users to search for images based on keywords. Similarly, the image is automatically generated. The new Chrome feature would give a descriptive text. For instance, it would say 'appears to be fruits and vegetables at the market'. The important note here is that Google is categorizing the feature as 'appears to be' to help the user understand it is generated by a computer and may not be fully accurate.

The Chrome feature with AI will be available only for those using screen readers with spoken feedback or Braille displays. The new description will be read out by the screen reader, but will won't show on the screen. Users can go to Settings in Chrome, select Advanced at the bottom to find the 'Accessibility' option. Here they can select 'Get image description from Google' option. Single web pages can also enable the option by right-clicking to open the 'context menu' and selecting the 'Get image description from Google' option.