According to Travis Leithead’s Windows blog post, Microsoft Edge will now automatically create alt text labels for images. If you don’t already know, these alt text tags are image descriptions so visually impaired users can understand what’s on the browser page when using screen readers.
Screen readers then synthesize the text into speech so that users can navigate even without seeing the contents of the screen. However, when this tag is not present, the player cannot provide a description to the user.
Now Microsoft Edge will automatically create labels using artificial intelligence:
When a screen reader encounters an unlabeled image, that image can be automatically processed by machine learning (ML) algorithms to describe the image in words and capture any text it contains.
Leithead says descriptions may not always be perfect at first, but having a description is already a big step towards improving accessibility for users.
The description of the images will be created through an Azure Computer Vision API which sends the images anonymously to the database, while guaranteeing user privacy. Autotags will not be added to images that are smaller than 50 x 50 pixels in size, images that are too large, or images deemed unsuitable for minors.
The novelty is now available for Windows, Mac and Linux and you can activate it in Settings > Accessibility in Microsoft Edge.
Additionally, Microsoft is already working to further improve the description of images that already contain this text using the same API, as they often do not contain all the information suitable for screen readers.