Web Accessibility: Audio and Video

Audio and video can be a great tool to get your website’s message to your users. In order for these mediums to be fully accessible to everyone, transcripts, captions and/or subtitles are required. Many people besides deaf or hard of hearing audiences can benefit from these aids, including (Penn State, n.d.):

Non-native speakers
When the vocabulary is unfamiliar
When the speaker is using an unfamiliar dialect
When the audio is poor or is not available.
When a person cannot listen to audio or video because of their work or study environment.

Per the Web Content Accessibility Guidelines:

If the webpage has audio, a text transcript or other text-based material should be provided (Penn State, n.d.).
If the webpage has video, captions or a text transcript should be provided (Penn State, n.d.).
1. Videos that include visual information critical to comprehension should include a description of the images or events for visually impaired audiences (Penn State, n.d.).
2. Any text displayed in a video, such as titles or captions, should comply with color contrast guidelines (Penn State, n.d.).
Video files should be embedded or displayed in a video player that can be accessed by a screen reader via keyboard commands. Examples include YouTube, Vimeo, QuickTime and RealPlayer (Penn State, n.d.). Kaltura, the video service supported by Purdue, is acceptable as well.
Do not auto-play video and audio files. It can be disruptive and jarring for people (Lambert, 2018).
Make sure to provide controls or options to pause or disable the video and/or audio (Lambert, 2018).

The next section will discuss transcripts, subtitles and captions.

Transcripts

Transcripts are a text equivalent of speech and are usually less expensive and time-consuming to create than captions or subtitles. They allow anyone that can’t access content from audio or video to read a transcript instead. In addition, scanning a transcript is much easier than skipping through a video to try to find information (Kalbag, 2017, p. 84).

Options

The least expensive way to produce a transcript is to transcribe the text yourself. This can be a slow and arduous process (Kalbag, 2017, p. 84). It also depends on the length of your media.
Professional transcribers are much quicker and can be good value for the money (Kalbag, 2017, p. 84).
Speech-to-text software can be used to dictate your transcript. However, such software works best in situations with only one speaker, since it relies on learning individual speech patterns (Kalbag, 2017, p. 84).

Tips

When transcribing audio or video as text, it’s helpful to indicate who is speaking, especially if there are multiple speakers (Kalbag, 2017, p. 84).
Include all of the relevant auditory and visual information in brackets to help the reader understand the context of the speech. Think of it like writing a script (Kalbag, 2017, p. 84).
Leave out irrelevant information, such as coughs or pauses, unless it adds context (Kalbag, 2017, p. 85).
Place the transcript below the video or audio. Add headings or links where applicable (Kalbag, 2017, p. 85).

Transcript examples:

Captions and Subtitles

If you’ve watched a movie in a foreign language, you’ve probably seen subtitles. Subtitles are lines of text that are usually translations of what’s being spoken on the screen. They appear at the same time as the spoken word, so you can understand the movie (Kalbag, 2017, p. 85).

Closed captions provide the same information, but also include other audio cues, such as “[music]” to describe when music is the only sound, or “[doorbell rings]” when an important sound effect is heard (Kalbag, 2017, p. 84).

Options

Producing your own captions will be the least expensive option and will enable you to write the most accurate text (Kalbag, 2017, p. 86).
Automatic captions are generated by machine-learning algorithms, so the quality of the captions may vary. For example:
1. YouTube has an auto-captioning option, but their speech-to-text software can be inaccurate if there is background noise, multiple speakers, or an accent unfamiliar to the software (Kalbag, 2017, pp. 85-86).
There are professional services that will create captions. These services can be expensive, as it is time-consuming work (Kalbag, 2017, p. 85).
1. Purdue works with a vendor called Cielo24 to provide mechanical or professional captioning services for videos uploaded to Kaltura. They offer a fast turn-around. Learn more about this service on Purdue’s Innovative Learning website.

Tips

Writing your own captions:

Captions are most comfortably read in short bursts. If there are pauses or silence in the speech, the captions can be shown for longer periods, giving readers more time to read and understand the text (Kalbag, 2017, p. 86).
Don’t show too much text on the screen, as that makes it harder to read alongside the video. Full sentences are too long, and the text shouldn’t carry over into two lines unless the screen size is very small (Kalbag, 2017, p. 86).
1. If you watch closed captions on TV or in a movie theater, you’ll get an idea of the right line length and where it’s comfortable to put a break in a phrase.
Only important audio cues that are necessary to understanding the content should be included in captions (Kalbag, 2017, p. 85).

Uploading captions:

Both YouTube and Vimeo have functions for uploading closed-caption files, and you can create closed captions for YouTube inside their video editor (Kalbag, 2017, p. 86).

YouTube: Add your own subtitles & closed captions
Vimeo: How to add subtitles and captions
Kaltura: Using subtitles and captions
- You may also email the Purdue Innovative Learning team for more information.

References:

Penn State. (n.d.). Video. Retrieved March 18, 2019, from https://accessibility.psu.edu/video/.
Penn State. (n.d.). Caption Guidelines and Policy. Retrieved March 18, 2019, from https://accessibility.psu.edu/video/captions/.
Lambert, S. (2018, April 9) Designing for Accessibility and Inclusion. Smashing Magazine. https://www.smashingmagazine.com/2018/04/designing-accessibility-inclusion/.
Kalbag, L. (2017) Accessibility for Everyone. Jeffrey Zeldman.