Accessibility has been a key focus of Google's for years, with the company always looking to add new ways for people to interact with their products so that they're accessible to as many people as possible. One of its standout accessibility features is Chrome's ability to caption images automatically when reading out a page, which is now available for more people in more languages.
In a blog post yesterday, Google explained why the Get Image Descriptions from Google feature is so important to people who are blind or vision impaired. There are millions of uncaptioned images on the web, and captions add much-needed context for users who use screen readers and brail devices. If those captions aren't available, things like tutorials can be harder to follow.
When Google added image descriptions to Chrome in 2019, it only supported English, with French, German, Hindi, Italian and Spanish being added in 2020. The company is expanding its reach further with the addition of Croatian, Czech, Dutch, Finnish, Indonesian, Norwegian, Portuguese, Russian, Swedish, and Turkish, bringing the total to sixteen supported languages. This was made possible by developing a machine learning model that can simultaneously translate the image caption into each supported language.
Google notes that its machine learning model still struggles with memes, sketches, cartoons, and screenshots, which is why Google has implemented an evaluation process to make sure the captions for these image types are correct before they're shown to users. It's encouraging to see Google continue to take accessibility seriously, and I'm looking forward to seeing this feature improve more over time.