Verify audio description (Prerecorded)
Description
This rule does not automatically validate anything. It simply reminds you to manually verify that prerecorded video content includes an audio description track when <video> or <audio> elements are detected on the page. Audio description (also called video description or descriptive video) is an additional narration track that describes key visual elements – such as actions, characters, scene changes, and on-screen text – that are not already conveyed through the primary audio.
Under WCAG Success Criterion 1.2.5 Audio Description (Prerecorded) (Level AA), all prerecorded video content must provide audio description so that people who are blind or have low vision can access the full content. Because an automated tool cannot determine whether the available audio description adequately covers all necessary visual information, a human auditor must verify both the presence and the quality of the description.
Disabilities impacted
- Blindness and low vision – Users who are blind or have low vision rely entirely on audio to understand video content. Without audio description, visual elements such as a character’s silent reaction, an on-screen chart, or a scene change are completely inaccessible. Audio description fills these gaps by narrating what is happening visually.
- Cognitive disabilities – Users with certain cognitive or learning disabilities may struggle to process rapid scene changes or complex visual information while also following dialogue. A well-paced audio description helps them build a complete mental model of the content without needing to divide attention between the visual and auditory channels.
- Motor disabilities – Users with motor impairments who rely on switch control, eye tracking, or voice commands may navigate video content in a non-linear fashion. Audio description ensures that visual context is available even when the user cannot easily adjust playback or re-watch segments to catch visual details they missed.
Why it matters
Video is one of the most common content formats on the web, used for education, entertainment, marketing, and communication. When visual information is not described, large segments of the population are excluded from fully participating. Providing audio description is not only a requirement under accessibility laws such as the ADA, the European Accessibility Act (EAA), and Section 508 of the Rehabilitation Act, but also improves the user experience for everyone – including people multitasking, non-native speakers, or those in low-bandwidth environments who rely on audio alone.
Coding problems and solutions
Common coding problems
- Missing audio description track – The video element contains no
<track kind="descriptions">child and no alternative version with audio description is provided. - Audio description track disabled by default – A
<track>element exists but is not configured to render as the default, and the video player does not expose a visible control to enable it. - Incorrect track
kindattribute – Usingkind="captions"orkind="subtitles"instead ofkind="descriptions", which prevents assistive technology from discovering the description track. - Incomplete coverage of visual information – The audio description narration omits critical visual elements such as on-screen text, data visualisations, or character actions, leaving the user without full context.
- Audio description out of sync – The description track timing does not align with the video playback, causing descriptions to play too early, too late, or to overlap with existing dialogue.
How to fix it
Use the <track> element with kind="descriptions"
The native HTML5 <track> element lets you attach a WebVTT file containing the audio description cues. Set kind="descriptions" and provide a descriptive label so users can identify the track.
<video controls width="640">
<source src="product-demo.mp4" type="video/mp4">
<track kind="descriptions"
src="product-demo-audiodesc.vtt"
srclang="en"
label="English audio description"
default>
</video>
<p>The demo includes an audio description
track that describes the user interface
elements shown on screen.</p><video controls width="640">
<source src="product-demo.mp4" type="video/mp4">
<!-- Missing audio description track -->
<track kind="captions"
src="product-demo-captions.vtt"
srclang="en"
label="English captions">
</video>
<p>A captions track is provided, but no
audio description track is available
for users who are blind.</p>Provide a separate video file with audio description
When the primary video cannot be modified (e.g., third-party content), offer an alternative version that contains an integrated audio description in the main audio track. Make the alternative clearly discoverable.
<video controls width="640">
<source src="interview.mp4" type="video/mp4">
<p>
<a href="interview-described.mp4">
Download the version with audio description
</a>
</p>
</video><video controls width="640">
<source src="interview.mp4" type="video/mp4">
<!-- No separate described version offered
and no track with kind="descriptions" -->
</video>
<p>Users who need description must find an
alternative version on their own.</p>Integrate audio description into the main audio track
For content created in-house, embed the audio description directly into the primary audio mix during post-production. This eliminates the need for a separate track and works on any video player.
// Post-production script that inserts audio
// description during natural pauses in dialogue.
// Actual description would be recorded and mixed
// into the audio file, not added at runtime.
const descriptionCues = [
{ start: 5.0, end: 8.5, text: "The presenter clicks the Settings icon in the top-right corner." },
{ start: 15.0, end: 18.0, text: "A modal window titled 'Preferences' slides into view." },
{ start: 30.0, end: 33.5, text: "The chart on screen shows a 40 percent revenue increase in Q3." }
];
// These cues are typically authored in a WebVTT file
// and associated via <track kind="descriptions">,
// not injected at runtime. The example above
// demonstrates the narrative content to include.<!-- The main audio track does not mention
any visual information at all. Users can
hear the dialogue but miss all on-screen
actions and text. -->
<video controls width="640">
<source src="tutorial-no-desc.mp4" type="video/mp4">
</video>
<p>This video relies entirely on visual cues
that are never described audibly.</p>Offer a programmatic toggle for description tracks
When using custom video players, expose a visible and keyboard-accessible button to toggle the audio description track. Use the TextTrack API to enable or disable the description track programmatically.
const video = document.getElementById('myVideo');
const descTrack = video.addTextTrack('descriptions',
'English audio description', 'en');
descTrack.mode = 'hidden';
const toggleBtn = document.getElementById('toggleDesc');
toggleBtn.addEventListener('click', () => {
if (descTrack.mode === 'showing') {
descTrack.mode = 'hidden';
toggleBtn.textContent = 'Enable audio description';
toggleBtn.setAttribute('aria-pressed', 'false');
} else {
descTrack.mode = 'showing';
toggleBtn.textContent = 'Disable audio description';
toggleBtn.setAttribute('aria-pressed', 'true');
}
});// Description toggle is hidden behind a
// right-click context menu that screen reader
// users cannot discover or activate.
video.addEventListener('contextmenu', (e) => {
e.preventDefault();
const menu = document.getElementById('hiddenMenu');
menu.hidden = !menu.hidden;
});
// Using kind="metadata" instead of "descriptions"
// so the track is invisible to assistive technology.
const wrongTrack = video.addTextTrack('metadata',
'Hidden description', 'en');Known Limitations
- Manual verification required – This rule flags
<video>and<audio>elements for review but cannot automatically determine whether the audio description is present, accurate, or complete. A human auditor must inspect the video content and its description track. This rule cannot catch issues automatically and depends entirely on developer or tester judgment. - Single-page applications (SPAs) – SPAs that dynamically insert video elements after the initial page render may not be captured if the selector runs before the SPA framework mounts the media elements. A second audit pass after the SPA has stabilised may be required.
- Shadow DOM encapsulation – Video players nested inside closed shadow roots are invisible to the DOM selector. Auditors must manually inspect all shadow roots where media elements may be embedded.
- Cross-origin iframes – Embedded video from third-party platforms (e.g., YouTube, Vimeo, Wistia) served inside cross-origin iframes cannot be inspected by this rule and require independent verification outside the scope of the automated scan.
- Dynamically loaded video sources – Videos whose
srcattribute or<source>children are set via JavaScript after user interaction may not be reflected in the DOM snapshot taken during the audit. - Audio description quality not assessed – Even when a
<track kind="descriptions">element exists, this rule cannot verify that the description content adequately covers all visual information. Human review of the actual WebVTT file or audio mix remains essential.
Resources
- Understanding WCAG SC 1.2.5 – Audio Description (Prerecorded)
- WCAG 2.2 Success Criterion 1.2.5
- MDN – The <track> element
- MDN – WebVTT API
- W3C WebVTT Specification
Rule
- Audit: accessibility
- Standard: WCAG
- Level: AA
- Success Criteria: 1.2.5
- Rule ID:
audio-description-prerecorded