By: Fady Wassef
In my short time at the DCP I have come to understand the meaning and necessity of accessibility for members of the disabled community and for a wide variety of cultural and ethnic groups. Lately I have become more aware of the need for captions for videos and other types of multimedia presentation. In recent years the rise of technology has not only been monumental in our society but also marked a cornerstone for future use of multimedia in education, employment, entertainment, retail and social media. A Nielsen poll disclosed that in 2018 online video accounted for 80% of overall internet traffic. Although media has become an ever present monument of our society, it seems that the desire to make it accessible to all has not collected the same amount of hype.
A problem that I have seen most prevalent on our campus is with bruincasted videos. Offering videos for students that may not even be comprehensible by some communities, in its nature, violates previous accessibility precedents. The ADA (1990) required that public spaces be made more accessible to those with disabilities. Although this didn’t specifically apply to online content or even the Deaf and hard of hearing community in general; it has been used in many cases to justify verdicts that are pro-captioning and pro-accessibility.
More specifically in the 2011 case of the National Association of the Deaf (NAD) v. Netflix the judge ruled that, “It would be ‘irrational to conclude’ that ‘places of public accommodation’ are limited to actual physical structures. It would severely frustrate Congress’s intent that individuals with disabilities fully enjoy the goods, services, privileges, and advantages available indiscriminately to other members of the general public.” This ruling set a precedent in applying the ADA not just to disabled communities and physical locations but rather interpreting it as legislation meant to be nondiscriminatory in essence. This specific case was one of the first times the law was applied in a broader sense and had a very positive implications for the Deaf community. The verdict fueled various different entertainment and education accessibility based cases in coming years. A conflicting ruling was in the 2011 case of Cullen v Netflix, where a deaf viewer sued Netflix on the same grounds as the NAD. The judge’s decision was the exact opposite however, saying that Netflix did not have to uphold the ADA since it was not a physical setting, this ruling however, was not published so was not intended to set legal precedent. I see this as a small but important first step into utilizing captions in the realm of accessibility.
Recently a handful of organizations and universities, both public and private, have been striving to make new media content accessible and in some cases remediating old content as well. Unfortunately these instances have conventionally stemmed from a lawsuit or investigation of sorts, yet still offer a very important historical precedent for media accessibility that we as a university must be at the forefront of. Additionally, some of these institutions are hesitant because of the cost and intensity of resources needed to transcribe or translate. To me this is all political, if these institutions cared enough about providing accessible content they would spend the money or find a cost effective alternative.
If I did not already make clear, the Deaf community faces a tremendous language barrier in terms of access to online multimedia content. Thus without a translator, any kind of video is rendered completely useless and unintelligible. The true value of captions is that they offer a way for the deaf and hard of hearing to easily have the same access to all the media content that we as a society use so pervasively. Obviously there are further language barriers concerning people who can’t read English, but captions can be easily translated into a wide variety of other languages. In this way, closed captioning can serve to reduce some of the discrimination felt by the Deaf and hard of hearing community by offering a means of accessing audible content in a transcripted manner. Additionally, captioning is cheaper, less intensive, and more sustainable than human translators.
Alright, that being said, what makes good captions? The Web Content Accessibility Guidelines (WCAG 2.1) has very specific success criteria for captioned videos to be considered accessible. That is a quality captioned video should be Accurate, Consistent, Clear, Readable, and Equal. Why are these elements important? For accuracy, if the transcription system or the captions themselves are not 100% accurate; there can be a range of minute to systematic errors that cause confusion and misinformation to the viewer. Captions must be consistent throughout the video and for other videos captioned in that style, this is crucial for short term and long term viewer understanding. They must be clear in that they have to provide speaker identification, and identification of music or any sound effects that are not explicit dialogue. Captions must also be readable, they must be synchronized to the video and be presented with enough time for viewers to read each line. On top of that they should not be obstructed by any of the visual elements. They must be equal in that the meaning and intention of the material is preserved. An example of this would be if someone said something really sarcastically and it was transcribed with no indication of sarcasm, then there might be some confusion for someone that is hard of hearing, reading the captions. Thus the words are the same but the meaning is changed and the captions would not be considered equal. So if you’ve ever wondered why platforms may have those brackets over the captions such as [sarcastically], [forced laughter], and in the case of the infamous Janice from Friends [machine-gun-fire like laughter], its to provide clarity on what is happening on screen. So, those little things with subtitles that we may get slightly annoyed by or take for granted sometimes are actually present to ensure equal access and understanding to everyone. Not to say I have not seen these descriptions fall short of the mark a couple times, like with [funky mystery music] and [intensity intensifies] but in general they are a useful and necessary tool. To highlight, quality captions must encapsulate the true meaning of the entire video, including sound effects and music, and be able to present it in a way that preserves that meaning.
It has been the DCP’s mission to make information accessible to all students, faculty, and staff. I believe that with a mission like that, excluding the Deaf and hard of hearing community would both be ignorant and embarrassing. That is why in the past few weeks we have been trying to stay ahead of the curve and have been testing an automated captioning program for Bruincast videos and other UCLA media content that is looking promising. The program is called Verbit! and compared to Youtube auto captions, which are powered by Google, it works much better. It offers a synchronized transcript of the video with an average of 99% percent accuracy. As we saw with other universities and organizations, a big concern was the cost and resource intensivity of providing quality captions, they simply could not afford it. However, with this program at 3 cents a minute and involving little to no human interaction, I personally would have a hard time understanding why it wasn’t looked into earlier. And although Verbit! doesn’t meet the success criteria for WCAG it is the absolute closest we can get to bridging the gap. AI is constantly evolving, so as time passes this system is only going to get better and better. Further the AI it uses can be trained to incorporate a glossary of terms such as scientific, cultural, and fictional terms, etc. It can also learn from past videos and course materials to provide higher accuracy. In other cases I have seen it nail videos dead on with 100% accuracy and punctuation. I have also seen the usefulness for this is program within the hearing community as it can help with more confusing terminology or professors who may be a little bit harder to understand. In essence this program is the best cost effective alternative to providing compliant quality captions and although it may not meet WCAG standards it would provide many students essential access to what is feasibly our most important web resource. With society ever advancing technologically we must make sure we don’t leave anyone behind.
Sources:
Originally Posted: 5 February 2020