A different kind of access need
Deafblindness describes a combination of hearing loss and vision loss. It does not always mean a total lack of both senses. Some people have a little residual hearing or sight, and others have none, and the combination can be present from birth or develop later in life. Because of this variety, there is no single way that deafblind people take in media. Communication preferences differ from person to person and may include speech, sign language adapted for the individual, tactile signing, braille, large print and others.
This article is general information. Anyone planning access for a specific person should ask that person, or an organization that works with deafblind people, what actually works.
Why on-screen subtitles often fall short
Standard subtitles assume the viewer can see the screen and read text on it. A person with enough remaining sight might do so with large, high-contrast text on a good display, perhaps from a short distance. But for many deafblind people, text drawn on a video image is not usable. It is a picture of words, and it appears briefly in a place the viewer may not be able to find.
So the aim for deafblind access is to get the words into a form the person can use, rather than to fit them into the picture.
Braille displays and similar devices
A refreshable braille display is a device with a row of small pins that rise and fall to form braille characters, which a reader feels with their fingertips. It connects to a computer, phone or other equipment and shows text as it is sent. If the text of the subtitles is available as data, in a separate file or stream rather than burned into the image, it can in principle be passed to such a device.
This is a main reason that how subtitles are delivered matters. Text that is part of the video image, as with burned-in subtitles, is hard for software to read. Text delivered as a separate track or in a transcript is far more flexible. Whether a particular setup works depends on the player, the device and the software involved, and testing with real users is the only reliable way to know.
Other formats may be used too. Some people prefer a transcript they can read at their own pace, in braille or in large print. For recorded material, a braille transcript can be produced in advance. Others may rely on a person who interprets for them using a tactile method.
Live situations
Live events are harder, since text has to be produced and delivered while people are speaking. Live captioning can produce a stream of text, and that stream might be sent to a braille display or shown on a large personal screen. Some deafblind people use a support worker or interpreter who conveys the content directly. The delay between speech and text is a real issue here, since a person reading by touch usually reads more slowly than a sighted reader. Caption delay may therefore be a larger concern than in other cases, and the person may need a way to catch up.
Planning is the key. Telling the organizers in advance, agreeing on the format and testing the equipment before the event usually does more than any particular technology.
Where audio description fits in
A deafblind person who cannot see the picture misses not only the sound but also the visual story. Audio description addresses what is on screen, but it is delivered as sound, which may not be usable either. In such cases the description can be written as text and delivered with the subtitles in a combined transcript or sent to a braille display. This idea of combining speech, sound information and visual description into one text document is sometimes the most practical route for recorded content.
What helps producers
If you make video and want to support deafblind access, a few habits help.
- Keep text separate from the picture. Provide subtitles or captions as a file or track rather than only burned in, so software can reach the words.
- Include non-speech information. Speaker names and important sounds should be in the text, since the viewer has no other source for them.
- Provide a full transcript. A transcript that includes description of key visuals can be read in whatever form the person prefers.
- Check accuracy. A person reading slowly by touch does not benefit from extra errors, so human reviewers are worth including for important material.
- Ask early. Needs differ, and a short conversation beats guessing.
The limits of what is known
Deafblind communities are small and varied, and research on media use is limited. It would be wrong to assume that a technology that works for one person will work for another. Support also varies a lot between places, as does the availability of equipment and trained interpreters. Where legal requirements apply, they differ by country and sector, so the relevant authority or a qualified professional should be consulted.
The broader lesson is a simple one. Subtitles are most useful when they exist as real text that can be moved into other formats. Treating them only as a visual layer closes off people who cannot see it.