Google first launched video captions eleven years ago in 2006. Three years later, in 2009, those efforts were taken to a whole new level with automated captions offered by YouTube. Fast forward to 2017, and the number of videos with automatic captions now exceeds a staggering 1 billion. Put differently, people watch videos with automatic captions more than 15 million times per day.
According to Liat Kaver, the product manager that made the announcement on YouTube’s Official Blog, a major goal for the team has been improving the accuracy of automatic captions — something that is not easy to do for a platform of YouTube’s size and diversity of content. He went on to say that a key factor in the success of this endeavor was improving their speech recognition, machine learning algorithms, and expanding their training data. All together, those technological efforts have resulted in a 50 percent leap in accuracy for automatic captions in English, which is getting YouTube closer and closer to human transcription error rates.
Of course the wide variety of content on YouTube makes it hard to caption everything, but it also offers the company with a wide range of training data. But the real shot in the arm comes from the YouTube video creators. themselves. When the YouTube community reviews and edits these automatic captions, that information then flows back into the machine learning process to make the system more accurate. As a result of all of these factors, the service’s automatic captions currently support 10 languages. And that is a major step forward for all of us who have to teach global audiences.