Microsoft has announced a new feature in the Edge browser designed to make it easier for people with visual impairments to interact with web content. We’re talking about automatic generation of image descriptions, which will be useful for visually impaired people who use screen readers to work on the computer.
Simply put, Edge can now automatically generate a description of the opened image, and the screen reader will play that text to the user. The Azure Cognitive Services cloud service is used when parsing and generating the image description. When the browser finds an untagged image, it pushes it to a Microsoft server for processing with the appropriate algorithm. The user does not need to make any efforts to generate descriptions for images.
The new feature works with many popular formats, including JPEG, PNG, GIF, and WEBP. Descriptions are currently supported in five languages. At the same time, the algorithm used to generate the description is able to recognize text from images in more than 120 languages. Microsoft warns that the description generation may not work perfectly at this stage, so in some cases the text will not be of very high quality. Also, descriptions are not created for images smaller than 50×50 pixels, large images, and content in some other categories.
The new feature has reportedly arrived in the latest version of Edge. To enable it, just go to the Options menu, go to the Accessibility section, and check the “Get image descriptions from Microsoft for screen readers” option. The feature is supported on Edge for Windows, Linux, and macOS.