You are here: joeclark.orgAccessibilityCaptioningBest practices in online captioning

Audio-described videos

It is unusual in any medium to find video with audio description that does not also have captioning. The latter is the more conventional, widespread, longstanding, and accepted accessibility practice, while audio description is viewed as something new. If any accessibility provisions are to be provided, in practice captioning usually comes first.

Sometimes audio description is difficult or impossible to provide in closed or optional format. In that case, the video may be open-described: Every viewer who can hear will hear the descriptions.

In other cases, description may be an optional feature called up by a different soundtrack (on DVD, on a set-top box, or in an online player); by Second Audio Program on a television; or by headsets in a DVS Theatrical screening. In these latter cases, audio description is closed, adding another soundtrack that needs to be dealt with.

Uncaptioned description

In the normal course of events, description is not captioned or even acknowledged. That will nearly always be the case even for closed description.

We recommend this approach in nearly all cases: Do not caption audio-description narrators. Description is mainly for blind people and captioning is mainly for deaf people. Those seem to be mutually-exclusive audiences, and that will actually be true in most cases. Deaf viewers won’t be able to hear the descriptions; even if they could, they can simply watch the video. Blind viewers won’t be able to see the captions; evne if they could, they can simply listen to the audio.

Audiences with multiple disabilities

But ambiguities pop up in the case of people with combined visual and hearing disabilities. Some low-vision people may also use captioning, for example. There is no available research on viewer preferences. But reading is slower than listening and is a mostly sequential process.

Thus it’s probably true that captioning dialogue, non-speech information, and speaker identification of a video production and also captioning the audio description will be too much to read. Also, viewers with enough vision to read captions may not need audio description.

Technical obstacles

Nonetheless, if authors wish to provide a captioned description narrator, they should be aware of what is technically achievable.

In online captioning, it is difficult to run two concurrent streams of caption text – particularly given the fact that captioning a description narrator will use different in and out points than captioning a program’s dialogue. (Narrators usually speak during pauses in dialogue.) If authors wish to provide captions of a description narrator, they generally won’t be able to run them simultaneously with program captions. If the different in and out times are deemed less important, the constraint then becomes the space available to caption that many lines – and viewers’ ability to read them.

It may be easier to digitize and provide separate files, one with captioning of dialogue, another with captioning of description narrator. Viewers could open either or both.