Google to Caption YouTube Videos
by Michael HoffmanThursday, November 19th, 2009
At See3 we have worked for organizations that have it in their mission to be accessible. For example, See3 client Easter Seals is all about helping people with disabilities. They need to model being accessible in their physical spaces as well as on their website.
See3 Director of Interactive Marketing and Fundraising, Shirley Sexton was the VP Interactive at Easter Seals for 8 years. She couldn’t do everything she wanted to do with video because videos playing in a Flash player — like those on YouTube — are not accessible. If you can’t see and hear it, it won’t work.
We have created workarounds for this within our own projects. And when asked about broader-based captioning what we have always said is that it’s a value, but it’s expensive. Someone has to create good captions before they can be added to the video.
We have been watching carefully the advances in speech recognition online with interest. People such as David Pogue are big believers in speaking into their PCs instead of typing. It didn’t seem a huge leap to us that someone would take this technology — which is amazing — and apply it to video. [Here’s a review from Pogue of this software from 3 years ago, and it is much better today than it was then.]
So it came as no surprise to us that Google is the one to step up. They have been leading advancements in all kinds of translation and other technologies, and oh, they also happen to own the biggest video site on the web, YouTube.
Here’s the lead from today’s NYTimes:
In the first major step toward making millions of videos on YouTube accessible to deaf and hearing-impaired people, Google unveiled new technologies on Thursday that will automatically bring text captions to many videos on the site.
While the technology can only insert captions on English language speech, Google is giving users the choice to use its automatic translation system to read the captions in 51 languages. That could broaden the appeal of YouTube videos to millions of other people who do not speak English but could use the captioning technology to read subtitles in their native language.
The speech recognition technology that Google uses to turn speech into text is not new; Google currently uses it to transcribe voice mail messages for users of its Google Voice service. But Ken Harrenstien, a deaf engineer who helped develop the automatic captioning system, said the technology had never been applied on such a large scale.
“This is some thing that I have dreamt of for many years,” Mr. Harrenstien said speaking through an interpreter. “To see it happen, is amazing.”




