YouTube Expands Video Captioning in 6 More Languages
Google's YouTube division is continuing its efforts to make it easier for viewers who are hearing-impaired or who speak different languages to view YouTube content using automatic captioning in their native languages.
YouTube has added support for six more languages—German, Italian, French, Portuguese, Russian and Dutch—to the automatic-captioning services that debuted in 2009, according to a Nov. 28 post by Hoang Nguyen, a YouTube software engineer, on the YouTube blog.
"Captions are important to make sure everyone—including deaf, hard-of-hearing, and viewers who speak other languages—can enjoy videos on YouTube," wrote Nguyen. "In 2009, you first saw a feature that automatically creates captions on YouTube videos in English, and since then, we've added Japanese, Korean and Spanish. Today, hundreds of millions of people speaking six more languages—German, Italian, French, Portuguese, Russian and Dutch—will have automatic-caption support for YouTube videos in those languages."
To use the service, viewers can click the red closed-caption ("CC") button on the task bar of a YouTube video and then select the language they want to use. Viewers can also choose to use a translation feature, which is presently in beta form, to translate the video's audio track into their native language.
So far, the automatic captions are available in 10 languages, wrote Nguyen, and are a good first step to expanding the program to provide high-quality captions for more videos on YouTube.
But it's not perfect, he wrote.
"As automatic captions will have some errors, creators also have several tools to improve the quality of their captions. Automatic captions can be a starting point, where creators can then download them for editing, or edit them in-line on YouTube. Creators can also upload plain-text transcripts in these languages, and the same technology will generate automatically synchronized captions."
So far, about 200 million videos on YouTube have automatic and human-created captions, and the number continues to grow, according to the post.
The automatic captioning of YouTube videos is only part of the story, however. In February, YouTube announced in a blog post that video owners who post their work could add captions and subtitles in some 155 supported languages and dialects, from Afar to Zulu.
In 2010, the Federal Communications Commission published rules governing closed-captioning requirements for video on the Web, according to YouTube, to ensure that hearing- and sight-impaired citizens could have access to video and audio on television, radio, online and through other media.
Andrew S. Phillips, a staff attorney for the National Association of the Deaf, lauded YouTube's efforts to expand the automatic-captioning services for its video content. Phillips, who is a deaf person, told eWEEK he uses audio captioning and it is helpful when viewing content.
"For a deaf or hard-of-hearing person watching video programs online, especially those that are consumer-generated, many programs are not required by law to be captioned," said Phillips. "But this tool, thanks to YouTube, improves accessibility to these programs giving us better access. It's nice to have the tools to do that and to help us when the law doesn't necessarily apply."
Ultimately, though, there are often problems with the accuracy of automatic captioning because it doesn't always translate correctly and can be affected by background noise, he said.
"Even if 90 percent of the information is correct, that 10 percent is a lot to still be missing or getting incorrect," said Phillips. "It's exciting, but until this technology is 100 percent accurate, we need people to add captions themselves to their videos."
The automatic-captioning service can also be beneficial for users who are not hearing-impaired, such as when a user finds a video on YouTube that is in another language. By using the captioning services with the translation feature, a user can understand a video made in another language, expanding the videos that they can view and comprehend.