Microsoft Edge is now capable of generating automatic images descriptions for users browsing the web with screen readers. The accessibility feature uses Azure Cognitive Services intelligence to generate alt text for web images that do not include it.
“When a screen reader finds an image without a label, that image can be automatically processed by machine learning algorithms to describe the image in words and capture any text it contains. The algorithms are not perfect, and the quality of the descriptions will vary, but for users of screen readers, having some description for an image is often better than no context at all,” explained Travis Leithead, Senior Program Manager, Microsoft Edge.
According to Microsoft’s data, more than half of the images processed by screen readers are missing alt text, which is a big problem for making the Internet more accessible. If Microsoft Edge can now provide automatic image descriptions for screen readers, some unlabeled pictures won’t be sent to Microsoft for processing: The list of exceptions includes excessively small (icon size or smaller) and large images, decorative images, and everything that Microsoft’s Vision API recognizes as gory, sexually suggestive, or pornographic.
Auto-generated image labels can be turned on in the edge://settings/accessibility menu on Windows, macOS, and Linux, and users will also need to turn on Screen reader support on the Accessibility Internals page at edge://accessibility. Windows 10 and Windows 11 come with a built-in Narrator feature, but other screen readers are also compatible.
Microsoft is planning to improve this new accessibility feature over time thanks to image recognition and algorithm improvements. “We’re excited for our screen reader users to benefit from this new service and grateful for the opportunity to work to improve the web experience for all users,” Leithead said today.