Automatic Captions in YouTube

Video has the potential to be a very stimulating and engaging format for teaching and learning, but at the same time staff have to be aware of the accessibility implications. YouTube continues to be the main player when it comes to web hosting of video clips and with 20 hours of footage uploaded every minute there is a vast resource of material. Despite the introduction of manual captioning by YouTube in 2006 the majority of this material isn’t full accessible. This might all be about to change as YouTube’s owner Google have announced automatic captioning for videos.

Using Google’s existing Voice speech recognition technology YouTube will be able to automatically add captions to uploaded videos.  Google point out that “the captions will not always be perfect, but even when they’re off, they can still be helpful—and the technology will continue to improve with time.”

As well as automatic captioning Google have also announced auto-timings. Auto-timing allows users to upload a text file with their video and the service will automatically work out when the words are spoken create captions for your video.

Before you go out and start uploading new videos for captioning you might want to wait to avoid disappointment as this service is going to be rolled out to these channels first: UC Berkeley, Stanford, MIT, Yale, UCLA, Duke, UCTV, Columbia, PBS, National Geographic, Demand Media, UNSW and most Google &YouTube channels.

Both these developments, whilst having specific accessibility issues in mind, should also improve general accessibility making it easier to search for videos using the automatic captions and timings to allow you to jump straight to the point in a video where a search term was said. A pilot for this was started last year with Google Audio Indexing.

Below is a video demonstrating automatic captioning on YouTube:

Click here for Google’s post on automatic captions in YouTube