Everyone who uses the social, video-sharing site YouTube may now be familiar with the automatic closed-captioning system—and mostly they know it by how funny it can be. For most of us, the automated closed-captioning is just a gimmick, something silly that cropped up on videos a year ago. We tried it once, and then promptly forgot it existed.

However, for the deaf community this sort of technology could be an amazing boon.

The only problem: the reason it was silly because it was so error prone it coughed up some ridiculously wrong transcriptions. Google, however, is getting to work on shoring that up, but it still looks like a long road. The Wall Street Journal Digits blog brings us the news on these developments,

In the past few weeks, Google has rolled out major improvements in the technology behind automatic captions, reducing the overall word error rate by 20%. The tool, which is accessed by clicking on the little “cc” button on most YouTube videos and selecting “Transcribe Audio,” can be used by anyone. But it’s not perfect.

YouTube doesn’t disclose numbers on the error rates because there is “quite a wide range” depending on the type of video, said research scientist Michiel Bacchiani, who heads up the speech recognition team for YouTube. “Things like one person speaking on a news broadcast will do quite well. If it’s a bunch of people yelling and running around with music in the background, it will do quite poorly,” he said in an interview with Digits.

Automatic-captioning would open up entire new vistas of accessibility to the disabled for interacting with video resources on the Internet as it takes away the need for a human to sit down and transcribe the video in order to caption it. This means more accessible media, faster—such as news broadcasts or even LIVE broadcasts. Even with a certain amount of error rate that would be better than absolutely nothing. It would still be up to actual humans to transcribe things for superior clarity, but it is a race between speed and resolution in this case.

In the article, Google’s lead caption software engineer, Ken Harrenstien hit the nail on the head when it comes to accessibility. “In general I would say the problem is awareness,” he says. “In developing a service, people are not thinking about how helpful captions are to so many people.”

Another excellent feature included with the automatic-captioning is an uploaded transcript that Google allows with YouTube videos. As a video is playing, a transcript can be scrolled past that follows along with the speech. This allows the video creators themselves to provide their own captioning—skirting the automatic-captioning, and making it more like closed-captioned television.

Another development from the uploaded transcripts happens to be better indexing for the videos. People who provide transcriptions, and thus captions, get more hits because people can textually search their videos for particular phrases and even get placed directly at that position in the videos.