Video is currently one of the most popular content publishing formats. Many users find video more engaging and more comfortable to follow than reading lengthy blocks of text. Most videos, however, convey at least as much substance through audio components as they do via visual elements. Care must be taken not to exclude the large number of users out there with hearing impairments. That’s where closed captioning comes into play.

Closed captioning has existed for several decades. It provides the text equivalent of audio content to viewers by superimposing text on the bottom of the screen in sync with the video that is playing. Hearing-impaired users can read the text captions while simultaneously watching the visual action.

A note about standards

In the early days of web accessibility, offering a transcript of audio content was considered “good enough.”  Now, however, transcripts are considered inadequate according to both the World Wide Web Consortium’s accessibility guidelines as well as most accessibility experts.

It’s important to provide users with video and audio content that’s synchronized. When a hearing-impaired user has to watch visuals within the video while also trying to follow a separate transcript, it’s hardly a great experience. So, although offering a transcript is better than nothing, closed captions are the accepted best practice. And, as we’ll see later, several shortcuts make adding closed captions much easier than it was in the past.

What not to caption

Videos that contain no audio, only music or only ambient noise do not need substantial captions. For example, if a video is silent, you can add a note to the video’s description (outside of the player). Or, add a caption to the beginning that reads “Silent Throughout” or something similar.

Likewise, if a video is set to music, add a note or a caption that reads “Music Throughout.” Even better, you might add details to the alert that read “Dramatic Music,” “Upbeat Music,” etc. Another example of a video that does not require major captioning would be a speech or presentation where a sign language interpreter is clearly visible within the frame.

Write captions that you would want to read

To see examples of how not to caption, turn captions on for a movie from the 1980s or earlier when watching Hulu or Netflix. The captions will likely be out of sync, riddled with typos, and lacking much-needed punctuation.

Would you want to read that?

Luckily, programming created over the last 10–15 years tends to be much better (not including computer-generated captions that accompany some live broadcasts which still tend to need more work). Creating quality closed captions is not mysterious. Correctly using punctuation and capitalization (but avoiding all caps) is a good start.

Experts will differ on how faithful to be to the audio content. For example, if a video depicts a performance where there is applause from the audience, that might be important enough to include in the captions.  Typically, non-speech audio is enclosed in brackets of some kind <applause> or {applause}.

You might choose to omit other ambient noises, however. If you’ve recorded a webinar and a bird was chirping incessantly outside the window, you should ask yourself whether that needs to be included. Try to exercise common sense and decide what would be necessary for hearing-impaired users to know. Maybe if someone in the webinar makes a joke about the bird, include it’s chirping in the captions. Otherwise, it probably will not help the viewer to know the bird was there.

Perhaps more controversially, you can also make judgment calls about actual speech. For example, if a webinar narrator spends 15 full seconds searching for words and stopping and starting over, you might decide to edit out some of the unnecessary speech. It depends on your viewpoint and the content you are captioning.

Design with captioning in mind

One of the biggest problems when it comes to closed captioning video is also the easiest to avoid. Closed captions appear at the bottom of the screen and usually take up the bottom 10–15 percent. If you’re shooting a video or designing slides that will become a webinar video, try to leave blank space at the bottom of the screen. Avoid adding logos, copyright information, or other content in the bottom 10–15 percent because that content will be covered when closed captions are superimposed over it.

Workflow shortcuts

YouTube and other video hosting services have introduced terrific innovations over the last few years to make creating quality closed captions easier. For example, if you have a script of the audio content for a video — which is quite common with webinar videos — you can upload it along with your video. YouTube’s artificial intelligence (AI) engine will generate closed captions from the script and even attempt to synchronize it with the visual content. Often, some tweaking and editing are needed. But this tool does a ton of the initial heavy lifting for you.

Alternatively, YouTube’s AI will also generate a set of automatic captions for any video uploaded. These captions are not published automatically, and should never be published without first proofing them. The AI will inevitably make mistakes, and some of these mistakes can be embarrassing. Still, the automatic captions do a lot of the grunt work for you. While editing and proofing are required, it’s a huge time-saver versus manually typing captions.

What if you don’t host your videos on YouTube or another hosting service? You can actually still piggy-back on YouTube’s technology. During the captioning process, YouTube also generates a captions file. To get to it, upload your video, edit the automatically-generated captions, and then download the finished captions file. You can then embed that caption file along with the video file wherever you ultimately want it hosted.