A few months back, I was fortunate to participate in a webinar hosted by Sonic Foundry that discussed the benefits of video accessibility and closed captioning, as well as the trends and legal requirements facing academic institutions.
The question and answer section of the webinar was very animated, with audience members bringing up thoughtful questions and critical concerns on a range of topics. In this blog post, I want to revisit the salient questions that were discussed and address a few that we didn't have time to get to.
You can also watch a recording of the full webinar here:
If a faculty member provides a clip of a full motion picture, is the campus legally allowed to caption that file?
Many people believe that this use case is squarely covered by Fair Use, which is a set of exceptions to the copyright law that balances the needs of content creators with those of content users. With closed captioning, Fair Use is evaluated on four factors. Another layer of defense is that Section 107 of American copyright law states that teaching is a purpose that is considered exempt from copyright infringement.
Can you provide a link to the source and more information about the BBC captioning study?
The BBC Office of Communications conducted a study showing that 80% of people who turn on captions do not have any hearing disability.
Will speech recognition technology advance to the point where humans are not necessary?
Speech recognition technology has made significant improvements in many different areas. The accuracy of speech recognition is affected by audio recording quality, accents, background noise, multiple speakers, complexity of the content, and the extent to which the speech recognizer has been trained. With academic video, the state of the art is about 80% accuracy. That means that one out of five words is wrong, and the unfortunate thing about speech recognition is that when it's wrong it's wrong spectacularly. The accuracy standard for closed captioning is 99%. While speech recognition will continue to improve, we are still decades away from reaching that level without human involvement. At 3Play Media, we use speech recognition as a starting point, but we also staff about 800 transcriptionists whose job is to clean up the mistakes left behind by the computer.
How does 3Play Media integrate with Mediasite?
3Play Media has a truly effortless captioning integration with Mediasite. You can submit closed captioning requests directly from your Mediasite account. The media files get automatically transferred to 3Play Media for processing and the closed captions get posted back to your Mediasite account after they have been processed. Standard turnaround is 4 business days, but you can select 2-day, 1-day, or even same-day service.
For accessibility compliance, is a transcript sufficient or do you need captions?
Transcripts are sufficient for audio-only content like a podcast. Captions are required anytime there is a video component, such as a recorded lecture or a slide show presentation with an audio track.
How do captioning vendors ensure high accuracy for complex or technical content?
At 3Play Media, we address technical and specialized content in a few different ways. First, our staff are highly trained on transcription and captioning standards. For example, we have specific standards on how to transcribe a mathematical equation or a chemical reaction. Our staff are also continuously audited to ensure consistent quality. Second, we make it very easy to upload terminology or other information along with your video. Adding names, places, or specialized vocabulary helps our transcriptionists to decipher what is spoken. Third, our staff cover a broad range of disciplines, and we try to match their expertise with your subject matter. Lastly, although rarely necessary, we provide an interface that makes it easy to edit captions after they have been processed. Any changes that you make automatically propagate to all the outputs and you don't need to reprocess your files.
Could you provide the references for the stats mentioned about disability in the world & US populations, causes of increased disabilities, etc.
Below are the sources cited in the webinar about disability trends in the world and the U.S.:
WHO World Report on Disability
Disability Employment Policy Resources
U.S. Census Bureau report on Americans with disabilities
U.S. Census Bureau Facts for Features
Can you talk more about how to get buy in to implement captioning?
Adding accessible measures is a shared responsibility. There are a number of parties that must work together to get buy in and make accessibility successful. Professors can be strong allies in advocating for accessibility. It's also helpful to form committees with individuals who passionately stand behind inclusivity. This is a great way to pool resources, knowledge, and connections to add momentum to policy efforts. Organizations like the National Association of the Deaf can also be helpful. This subject is covered in depth in the white paper Roadmap to Web Accessibility in Higher Education.
Are there any creative ways to pay for captions?
We've compiled a list of resources showing how some of our customers pay for captioning through grants and other sources of funding.
What are the considerations for captioning in-house versus using a third party vendor?
This is a complicated question with many different considerations covered in depth by Korey Singleton from George Mason University in the webinar In-House Captioning Workflows and Economic Analysis.
How does the new standard for caption quality affect academic video and how does it affect video platforms?
The new FCC caption quality standard released in 2014 specifies caption accuracy, synchronicity, completeness, and onscreen placement. Although these standards directly apply to broadcast television, they extend to online video due to the CVAA law passed in 2010. Under the CVAA, the copyright owner has the responsibility of providing captions. Video platforms are responsible for allowing the captions to pass through to be rendered on the video. CVAA and FCC rules only apply to academic video to the extent that the content airs on television. In most cases, Section 508 + 504 and the ADA are the laws that dictate captioning requirements for academic video.