Windows 10’s Narrator uses artificial intelligence to automatically write photo captions

Narrator is a tool in Windows that reads text on your PC screen aloud and describes events, such as notifications or calendar appointments, so you can use your PC without a display. Visually impaired people use this feature to interact with their PCs. When you open a photo, Narrator will read the alternative text available for the photo. As you know, most of the photos doesn’t include an alternative text making it difficult for visually impaired to consume the photos. Microsoft is now trying to solve this issue using artificial intelligence.

Yesterday, Microsoft released Windows 10 Build 16226 for PCs to Windows Insiders. This build includes an updated Narrator with auto-captioning feature. Narrator will use artificial intelligence to generate descriptions for images that does not include an alternative text. It can also now extract text from images using optical character recognition. This feature will change the way visually impaired people consume photos on their PCs.

To try this feature, turn on Narrator, set focus to an image and then press Capslock + Shift + D to hear its description. As it makes use of Microsoft’s Cognitive Services APIs, you will need to be connected to the internet for it to work.