YouTube extends automatic video captioning

YouTube users have been able to manually add captions to videos since 2008 and in November of last year the site began offering machine-generated captions for about a dozen partner channels.

Hiroto Tokusei, a YouTube product manager, said in a blog post on Thursday that the automatic caption, or auto-caption, feature was now being expanded to all videos on the site in English.

Auto-captioning uses speech-to-text technology to generate subtitles.

"Making video easily accessible is something we're working hard to address at YouTube," said Tokusei, citing studies that predict that over 700 million people worldwide will suffer from hearing impairment by 2015.

The auto-captioning feature will make YouTube videos "more accessible to people who have hearing disabilities or who speak different languages," the YouTube project manager said.

While the auto-caption feature currently only works for videos where English is spoken, Tokusei said users can use Google's automatic translation service to simultaneously translate the captions into 50 other languages.

Auto-captioning in more languages will be added "in the months to come."

The YouTube project manager noted that auto-captioning is not yet perfect and a "clearly spoken audio track" without background noise is needed to create quality captions.

Video owners do have the capability, however, of downloading auto-generated captions, improving their accuracy and then uploading a new version.

One of the chief advocates for captioning capability at YouTube has been Vint Cerf, the Google vice president who has been described as the "Father of the Internet."

Cerf, who is hearing impaired and has been wearing hearing aids since the age of 13, made a personal appearance at the unveiling of the YouTube auto-caption features at Google's Washington offices in November.

(c) 2010 AFP