By means of translating the codes of a photo or what is called as Facebook’s neural networks into alternative text, screen readers are able to provide a more detailed description of what a photo is about like “image may contain: three people, smiling, outdoors.” Take note that Facebook is using the phrase “may contain,” which means what could be described is not hundred percent accurate.
Prior to this feature, screen readers can only read text and say “photo” when there is an image included. Facebook users who are blind have to depend on their friends and family members to describe what the photo contains.
Right now the availability of this new Facebook feature is limited to the English language in US, UK, Canada, Australia, and New Zealand and in iOS devices. However, Facebook plans to make automatic alt text available in other languages and platforms in the coming days.
