Accessible Video Design: Captions, Controls and Testing

Accessible Video Design

Somebody is watching your video on a train with the sound off. Somebody else is deaf. Somebody else has the volume up and cannot follow the accent.

All three need the same thing, and only one of them is who most people picture when accessibility comes up. That is the useful way into accessible video design: the features exist for a specific group and get used by almost everybody.

Accessible Video Design

1. Video Captions and subtitles are not the same thing

The terms are used interchangeably, and they describe different jobs.

  • Subtitles assume you can hear the audio and need the words in text, usually for translation or clarity of accent. They carry dialogue only.
  • Video Captions assume you cannot hear the audio at all. They carry dialogue plus everything else conveying meaning: who is speaking, relevant sound effects, music where it matters, and significant silences.

A video with subtitles is not accessible to a deaf viewer, because “[door slams]” and “[laughs]” are where a substantial part of the meaning lives. If you are meeting an accessibility requirement, you need captions.

2. Auto-generated captions are a draft

Automatic captioning has improved considerably and it is still a draft.

It struggles with proper nouns, technical vocabulary, regional accents, overlapping speech and background noise. In a business video that means your company name, your product names and your industry terms are precisely what it gets wrong.

Review every one. It takes a fraction of the time writing them from scratch would, and unreviewed captions frequently say something different from what was said, which is worse than none because viewers trust them.

Check punctuation and speaker identification too. A wall of unbroken text is technically captioned and practically unreadable.

3. Provide a transcript as well

A transcript is the full text as a separate document or page section.

It serves people who would rather read than watch, people using screen readers who find video controls awkward, and anybody wanting to search the content or quote from it.

It is also the cheapest accessibility feature available, since you already have the text from the captions.

4. Accessible video controls people can actually use

Video players are a common accessibility failure because custom players routinely rebuild native controls badly.

Every control needs to be reachable and operable by keyboard, with a visible focus indicator. Play, pause, volume, captions toggle, fullscreen, and the scrubber.

Never autoplay with sound. Beyond being disruptive, it interferes with screen reader output and can be genuinely distressing.

Give people the ability to pause anything that moves. WCAG 2.1 Level AA requires a mechanism to pause, stop or hide moving content lasting more than five seconds.

WCAG 2.1 AA infographic of each level.

5. Audio description, where it applies

If your video conveys information visually that is not in the audio, somebody who cannot see the screen misses it.

Sometimes the fix is simple: have the narration describe what is being shown, which serves everybody and needs no separate track. This is worth planning at the scripting stage rather than solving afterwards.

Where that is not possible, a separate audio description track or a detailed text alternative fills the gap.

6. The weight question

Video is the heaviest thing most sites carry.

Never autoplay it. Load it on interaction rather than on arrival, using a poster image as the placeholder. Serve appropriate resolutions rather than one large file for every device.

Where performance and environmental impact both apply, treat them as the same lever: reducing unnecessary data transfer and processing supports a faster experience while lowering estimated digital carbon, though environmental figures remain modelled rather than directly measured.

A video that demonstrates something photographs cannot earns its weight. A looping clip behind a headline does not.

7. Testing accessible video design

Watch it with the sound off and see whether you can follow it. Then operate every control with the keyboard alone. Then check the captions against what was actually said.

Those three checks take ten minutes and find most problems.

Our sustainable web design work treats media accessibility and media weight as one decision.

Most common questions

01

What is the difference between captions and subtitles?

Subtitles carry dialogue for viewers who can hear the audio. Captions carry dialogue plus speaker identification, sound effects and other meaningful audio for viewers who cannot. Accessibility requirements need captions.

02

Are automatic captions good enough?

As a first draft. They reliably struggle with proper nouns, technical terms and accents, which in a business video means your company and product names. Review every one, since inaccurate captions mislead viewers who trust them.

03

Do I need a transcript as well as captions?

It is worth providing. Transcripts serve people who prefer reading, people finding video controls awkward, and anybody searching or quoting the content. Since you already have the caption text, it costs almost nothing.

04

What does accessible video design require of a player?

Every control reachable and operable by keyboard with a visible focus indicator, no autoplay with sound, and a way to pause moving content. Custom players frequently rebuild native controls badly, which is where most failures originate.

05

When is audio description needed?

When the video conveys visual information not present in the audio. Often the simplest fix is writing narration that describes what is shown, which serves everybody and avoids needing a separate track.

06

How do I stop video slowing the page?

Never autoplay, load on interaction with a poster image rather than on arrival, and serve appropriate resolutions per device. Video is usually the heaviest element on any page carrying it.