You’re streaming a video. Bright yellow text flows across adark screen: a movie title followed by a long-ish narrative. A spaceship fliesacross the screen, chased by a much larger one; colored laser lines streakacross the screen, missiles fired by the larger craft at the fleeing spaceship.The scene then moves inside the first ship, where three robots hurry along ahallway; armed soldiers rush past them.
Finally, more than two minutes in, one of the robots speaks thefirst line of dialogue.
It’s the opening of StarWars, of course. But if you couldn’t see what was on the screen, you’d becompletely confused. Obviously, something needs to be done to make the movieaccessible to people with low or no vision. But what?
When using a medium like video, learners need to know bothwhat is being said and what is showing on the screen. Most people are familiarwith captions and transcripts, which provide access for learners who arehearing impaired or deaf. But what about learners with impaired vision? Canvisual media, including video-based eLearning, be accessible to learners withlow or no vision?
Though less familiar, the solution is equally essential:audio description. “It’s not like captioning, where everybody sees it on thetelevision in the bar or in the gym,” said Joel Snyder, president of Audio Description Associates anddirector of the American Council of the Blind’s Audio Description Project. “With description, you’re not aware that it’s thereunless you’re using it.”
What is audio description?
Audio description fills the visual void with a narrateddescription of the scene, the characters, and the action, providing access tovisual elements that are not described in the default audio. The description isprovided between lines of dialogue and other sounds on the soundtrack; it mightbe detailed or quite brief, depending on the content and space available.
Audio description can be provided for live or recordedcontent; some theaters and concert halls offer the service, as do many museumsand galleries. Some television and movie content is available with audiodescription—including all seven Star Warsmovies. Audio description of recorded content generally omits decorativeelements, such as clip art or icons that do not add new information.
The audio track with description is often a separate audiotrack that users can select in the same way as an alternate language track orvia an accessibility menu. Viewers who activate the audio description option ona video player or streaming service will hear the augmented track; otherviewers will hear the default audio, sans description.
Description is not the only option
For some types of content, another approach is possible:designing content, particularly eLearning, to be accessible from the ground up.
“Universal design is all about one-size-fits-all; basically,none of this separate-but-equal,” Snyder said. For example, an audio tour at amuseum can be accessible using either approach.
“A regular audio tour, if you will, provides a lot ofinformation—background, facts, and such,” Snyder said. “That same tour can bewritten so that it provides descriptive material.” When done well, the addeddescription will not seem “odd” to sighted museumgoers, Snyder said, but thetour is more accessible—to all hearing museum visitors. “Some museums end uphaving two separate tours, which seems a little silly to me. It’s unnecessary,especially when people talk about universal design,” he said.
eLearning and audio description
Likewise, some types of eLearning, such as short slide-basedpresentations, offer two paths to accessibility.
- The presenter speaks aloud all the text that appearson the screen, along with detailed explanations; she also describes aloud allcharts, graphs, or photos. The resulting short video is accessible, as-is, tolearners who have visual impairments. All the relevant information is in thedefault audio.
- The presenter, as is all too common, does notdescribe the images or even read all the text that appears on screen. Thisvideo would need audio description to be accessible to learners with visualimpairments.
“I encourage people to do self-description, which is not‘I’m six-foot-five and I have black hair.’ Self-description is providing thedescription yourself, during your presentation,” Snyder said. “Similarly, ifyou’re going to present a video, it should be audio described, and it should becaptioned.”
Other types of eLearning content present a greaterchallenge. For instance, consider an interactive activity that presentsinteractions among several characters. The learners’ task is to resolvecommunication problems. In one scenario, a character strides out of the room.Another puts on a headset and deliberately—and silently—turns away from thegroup. Here, the characters’ actions, facial expressions, and body language areall relevant to learners’ understanding of the group dynamics. Without an audiodescription, learners with visual impairments would miss critical elements ofthe content.
What about a five-minute training video that walks learnersthrough the steps of a process? Ideally, the default audio would include anaudible explanation of each step, rather than assuming that all learners willunderstand what they are watching the actor do. If thorough explanation is partof the default audio, audio description is not needed.
Description aids understanding
Federal law requires that some web content and eLearning beaccessible, including captioning, descriptive “alt” tags on images—and audiodescription. (See “Section 508 Refresh: The Clock Is Ticking on eLearning Accessibility Requirements”for information on these requirements.)
But beyond legal requirements, description of the visuals ineLearning content is simply good design. “Any image that you project, any imagethat’s up there, whether there’s an alt tag or not, should be described in textsomewhere,” Snyder said. “Going back to the question of universality—peopleseem to learn in different ways; that makes it more accessible to anyone. Theyhear as well as see the particular image. That’s a lot of what eLearninginvolves, I think.”
Research indicates that the majority of people who usecaptions are not deaf or hard of hearing. They use captions to help them focus,for example, or to better understand the material. (See “Improve Engagement, Focus, and Comprehension with Closed Captions for eLearning Videos.”)
Snyder says that’s true of audio description as well.“Sighted people find that they notice things that are pointed out to them. Youmight go to a film, you like it, and you go to see it again. And, ‘Oh, whoa, Ididn’t see that the first time.’ Well, you probably would have seen it thefirst time, had the audio description been on.”
Description might also aid learners in ways that go beyondthe specific content: “There’s increasing evidence that description aidsliteracy, in the same way that captioning helps kids or speakers of otherlanguages to learn English. Hearing varied word choice, hearing synonyms,hearing similes and metaphors helps you develop a sense of literacy. In thatway, it’s helpful for everybody,” Snyder said.
Blind Spot blogger Hannah Thompson’s eloquent post, “Audio Description,” makes it clear that all video-watchers could benefit from theadditional narration: “Last night I watched a film for the first time in mylife. I have been to the cinema hundreds of times and watched thousands ofvideos and DVDs, but yesterday I realized what watching a film reallymeans. … Hearing a little bit of extra detail as I was watching the filmwas a hugely enriching experience. It was unobtrusive, informative, andengaging. And it made me realize that there are … elements of film that Ihave been missing.”
A three-step process
The three steps in creating audio description use differentskill sets and are not necessarily performed by a single developer.
1. Writing a script
The script is a description of thescene, the characters, the action—everything except the dialogue andother sounds. The describer should include details that enhance understanding:mentioning what a character is wearing or eating; describing the flash of angerthat crosses an actor’s face or the care he’s taking to hang a photo just so. Descriptionof items for a site tour or art gallery tour should mention what a sculpture ismade of, the colors that dominate in a painting, the mood a photo creates.Other salient details could include placement of items relative to the listeneror to walls, doors, or other objects; items’ size; and description of thelighting and environment. The describer has to strike a balance betweenproviding relevant information and overwhelming the listener.
2. Voicingthe description
This requires recording each lineof the script, with pauses between lines. Depending on the project,professional voice talent might be used—or an audio describer can both writeand voice the script.
3. Editing the audio
Any video editing tool that allowsthe editor to add and edit audio tracks, or a multiplexing/demultiplexing tool,such as Subler for Mac, can be used to integrate the audio description with thedefault audio.
The soundtrack takes precedenceover the audio description, which is integrated into pauses in the defaultaudio. “Hopefully, the description fades into the background; it doesn’tdominate the experience,” Snyder said. Description does not have to syncperfectly with the action in the video, but the description must be closeenough that it remains relevant.
Adding audio description is “really not complicated at allif you’re talking about just slides or images. It depends on the eLearningcourse, what the structure is,” Snyder said, and many developers do itthemselves. For content that is heavy on videos, “You generally want to have aprofessional look at the video, listen to the video, and create descriptionthat fits within the pauses between bits and pieces of dialogue or criticalsound elements.”
For companies that produce large amounts of eLearningcontent, Snyder suggests a training session for developers. “There arefundamentals to it; it’s not just labeling. There’s a way to do descriptionthat represents best practices,” he said.
Audio Description Associates and other audio description companiesoffer training that ranges from a few hours to several days. Snyder’s book, The Visual Made Verbal: A ComprehensiveTraining Manual and Guide to the History and Applications of Audio Description,also provides guidance.
