Deprecated: Assigning the return value of new by reference is deprecated in /home/cather20/public_html/blog/wp-includes/cache.php on line 36

Deprecated: Assigning the return value of new by reference is deprecated in /home/cather20/public_html/blog/wp-includes/query.php on line 21

Deprecated: Assigning the return value of new by reference is deprecated in /home/cather20/public_html/blog/wp-includes/theme.php on line 540
l’azile » between the lines

April 2, 2011

between the lines

catherine @ 9:48 pm

Update: Universal Subtitles now goes by the name of Amara. However, the links to the website contained in the following post are still functional. (2013/02/06)

A little while ago, I captioned a video at Universal Subtitles, an outreach video about reserved parking for people with disabilities (in French). The video calls people out on not respecting handicap parking spaces and dispells some myths (eek!, did I really say that?) about the real consequences of not considering why these things are the way they are.

And actually, I started this project before Christmas but got sidetracked and it took me several, euh, I was going to say weeks but I guess it is months… So anyway, it took me a few months to find time to do this. Because the truth is that you really need to set aside some time for this activity, however you want to categorize it. I mean, unless you are a professional captioner, this will take some time. Have I repeated that enough for you? Have I? Because this project took me several hours. I had to learn the system, revise the document and then had to actually hunker down and do the transcription, synchronization, revision, stressing out on details, etc.

All that being said, the folks at Universal Subtitles make it really easy. The tutorials are simple and easy to follow, the captioning interface is, again, as easy as possible and user support is superb. Indeed, Dean, who answered many of my questions, was the epitome of the 3 P’s: polite, professional and patient. Bravo Dean, you made it through!

And I think Universal Subtitles can become a great crowd sourcing tool for captioning videos on the Web on a shoestring or totally non-existent budget. I can truly imagine awesome volunteers dedicating a few hours here and there to caption videos for non-profits because many non-profits (or other cash-strapped organizations) do not necessarily have the means and/or know-how to caption video resources. So this can really help and I hope people will experiment and utilize this tool.

This whole exercise got me thinking about captioned Web resources in general and how the number of videos out there is staggering and can lead one to believe we are fighting a losing battle when it comes to making all these resources accessible to the Deaf and people with other hearing impairments. And it really is mind-boggling when you think about it. According to this article on mashable, 65,000 videos were being uploaded to youtube per day in 2006. And this May 2009 Digital Media Wire article states that 20 hours of video are uploaded every minute! Those numbers are a bit dated of course and they do not account for all the other resources, like Vimeo, etc., not to mention corporate sites and official online video streaming from television networks, but it gives you an idea of just how huge the challenge is.

Some in the mainstreaming crowd will point out that captions can gain traction by their usefulness to people without disabilities. For example, more and more televisions in public places such as sports bars, airports, waiting rooms, etc., have captions turned on to make content more understandable by whoever happens to be there. And they are right, it is a nice example of auxiliary benefits that everyone can understand without stretching the truth or making wild comparisons.

The thing is though that television and video resources on the Web (or even that other beast, Web television) are, for the moment anyway, very different animals. So while captioning on television is *cough* strictly regulated in some countries, like Canada, and there are *cough* solid ways to enforce these regulations, the reality, regardless of regulation, is quite different on the Web.

So, while this is not my area of expertise, like many of my peers, I think automation is our best bet in trying to make some headway with regards to captioning the volume of content that is produced (though I doubt we will ever get ahead of the wave, not in my lifetime anyway, and I would truly love to be proven wrong). Several noteworthy efforts in this area have yielded some promising results, namely, of course, Google’s introduction of “auto-caps” for youtube. OK, it is not perfect but surprisingly not too bad in many cases as well as a clever way of avoiding a lawsuit. Others are also experimenting in this area, such as a local research center, CRIM, who has been working with speech recognition for captioning for a while now, work which has resulted in the the spin-off of a specialised firm, SOVO. There are other efforts, some with a mix of automation and human “clean-up crews”, if you will and I am sure you can offer other examples.

Will automation ever be able to replace, at equal value, captioning done by humans? Personally, I do not think so. At least, again, not in my lifetime though, again, I would love to be proven wrong. I mean, captioning itself is, unfortunately, often a poor substitute for the real thing, just as audio description is. And yes, I know how that sounds but anyone who has heard or seen before and then had to rely on the related accommodation knows what I mean. Which is not to say it is not worth it. It totally is.

However, automated captions rely on machines getting the mechanics, the basics, right, which they often do. But can a machine pick up a change, subtle or not, in tone? Can it understand the difference between a sigh of relief and one of exasperation? And I will not even get into ambient sounds, which are also key to not only figuring out what is going on but appreciating it for a multitude of people with sensory impairments. I hope someday computers can really perform in this area because it will be a great win for people with disabilities. In the meantime, for quality work, you can not beat humans. Yay, go humans! And automation is a great way of “making a dent” :)

Anyway, so here is that video I captioned (in French and with a strong Québécois accent!). Enjoy.

PS. Metadata is also an important key. More on that some other time.

1 Comment »

  1. Brilliant film and beautiful captioning! As a person with autism and auditory processing disorder, I rely on captions to fully understand the words being said by people on TV (in person, I find that my slight lip reading skills help, but on TV not that much). So many services on the internet do not have captioning. Streaming video on Netflix and Amazon Video on Demand/Instant Video comes to mind- I’m guessing it has something to do with the technology and distribution/copyright, but I’m not sure, though I do know there has been a great deal of user discussion over on Amazon.com’s community forum about it. Anyway, I think you did a fantastic job captioning this. I have not really spoken French since my 2nd year of high school and, with the knowledge of what the general topic would be and the visuals, I was able to really understand what was going on and the important dialog between the two able bodied people was clear to me. Big kudos to you for a job well done and for your public service!

    Comment by Bek — May 2, 2011 @ 4:25 pm

RSS feed for comments on this post.

Leave a comment

XHTML (You can use these tags): <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <code> <em> <i> <strike> <strong> .

Please note that first-time comments will be held in moderation. For more information, see moderation policy.


link rel=