WCAG 1.2.7: Extended Audio Description (Prerecorded)
Standard audio description has a hard constraint: the narration must fit into the pauses between dialogue. When a video never stops talking, those pauses do not exist. This AAA criterion says that in that case, the video itself must pause so an extended description can convey what is on screen, then resume.
The success criterion, in full
Where pauses in foreground audio are insufficient to allow audio descriptions to convey the sense of the video, extended audio description is provided for all prerecorded video content in synchronized media.
The conditional matters: extended description is only required where pauses are insufficient. If ordinary gaps in the soundtrack are long enough for standard description to convey the sense of the video, that standard description satisfies this criterion too.
AAA context: the audio description ladder
Three criteria form a progression. 1.2.3 (Level A) lets you choose between an audio description or a full text alternative. 1.2.5 (Level AA) requires the audio description itself. 1.2.7 (Level AAA) closes the last gap: content where even a well-written standard description cannot fit, because the foreground audio leaves no room.
Think of a software walkthrough where the presenter talks continuously while clicking through menus, never naming what they click. A blind viewer hears an unbroken monologue and misses every action. Freezing the frame — “The presenter opens the Settings menu and selects Privacy, third item in the list” — and then resuming is the only way to make that video genuinely equivalent.
Who this helps
People who are blind
All essential visual information — actions, on-screen text, scene changes, charts — is narrated at sufficient length instead of being compressed into gaps or dropped.
People with low vision
Viewers who can see the video but not read on-screen details (menu labels, code, small diagrams) get those details spoken in full.
Learners using audio only
Anyone consuming instructional content without watching the screen — a common situation for training material — can follow every demonstrated step.
People with some cognitive disabilities
Pausing the action while the visuals are explained gives extra processing time, which the Understanding document explicitly calls out as a secondary benefit.
How extended audio description works
- 1Identify the points where essential visual information appears but the foreground audio leaves no usable pause.
- 2At each point, freeze the video (and hold or duck the programme audio).
- 3Play the extended narration describing the visual content — as long as it needs to be.
- 4Resume normal playback. Repeat wherever needed; the described version ends up longer than the original.
✓ Passes
- A second, clearly linked “extended described” version of a dense tutorial, with pauses edited in for full narration.
- A player that pauses at scripted cue points, speaks a description, then resumes automatically.
- A dialogue-heavy video whose natural pauses happen to be long enough — standard description conveys the sense, so no extension is needed.
✗ Fails
- Standard description that skips essential visuals because “there was no room” in the soundtrack.
- Description narration that talks over the programme dialogue, making both unintelligible.
- An extended version that exists but describes only some scenes, leaving key demonstrations unnarrated.
Implementation patterns
1. Link an extended-described version
Produce the extended version in your editing tool (freeze-frames plus narration) and offer it alongside the original. This is the most robust approach because it works in every player.
<video controls src="/media/dashboard-tour.mp4">
<track kind="captions" src="/media/dashboard-tour.vtt"
srclang="en" label="English" default />
</video>
<ul class="media-alternatives">
<li>
<a href="/media/dashboard-tour-extended-ad.mp4">
Extended audio described version (runs 9:40 instead of 6:15)
</a>
</li>
</ul>2. Script the pauses in the player
For a single-source workflow, define cue points where the player pauses, plays a description clip, and resumes. The same cue data can generate a text transcript for 1.2.8.
const cues = [
{ at: 42.0, clip: "/media/desc/settings-menu.mp3" },
{ at: 95.5, clip: "/media/desc/import-dialog.mp3" },
{ at: 180.2, clip: "/media/desc/final-chart.mp3" },
];
const video = document.getElementById("tour");
let next = 0;
video.addEventListener("timeupdate", () => {
if (next >= cues.length) return;
if (video.currentTime >= cues[next].at) {
video.pause();
const desc = new Audio(cues[next].clip);
desc.addEventListener("ended", () => video.play());
desc.play();
next += 1;
}
});
// Reset the cue pointer when the user seeks backwards
video.addEventListener("seeked", () => {
next = cues.findIndex((c) => c.at > video.currentTime);
if (next === -1) next = cues.length;
});3. Make the option discoverable
However it is delivered, users must be able to find and enable it — a labelled control near the player, not a buried settings page.
<button type="button" id="ext-ad-toggle" aria-pressed="false">
Enable extended audio description
<span class="hint">(video pauses for longer descriptions)</span>
</button>How to test for 1.2.7
- 1
Watch the video with the screen off (or eyes closed)
Note every point where something essential happens visually that the soundtrack does not convey — actions, on-screen text, charts, scene changes. This is the information description must carry.
- 2
Assess whether natural pauses are sufficient
Compare the amount of essential visual information against the available gaps in the foreground audio. If the standard description track already conveys the sense of the video, 1.2.7 is satisfied without extension.
- 3
If pauses are insufficient, confirm an extended version exists
Look for a linked extended-described version, a player mode, or scripted pauses. Its narration must cover all the essential visual information you catalogued.
- 4
Play the extended version end to end
Verify the video actually pauses for the added narration, the descriptions are accurate and complete, and playback resumes correctly — including after the user seeks.
- 5
Check discoverability
A user who needs extended description must be able to find it from the original video without guesswork: an adjacent link or a clearly labelled player control.
Common failures
- Dropping essential visual information from the description because the soundtrack had no pauses — the exact situation extended description exists for.
- Cramming rushed, truncated descriptions into gaps that are too short, producing narration nobody can follow.
- Describing over the top of dialogue instead of pausing, so users lose both the description and the speech.
- Publishing an extended version but never linking it from the original video page.
- Scripted player pauses that break when the user seeks or replays, so descriptions fire at the wrong times or twice.
- Treating a text transcript as a substitute — a transcript can satisfy 1.2.8, but it does not satisfy 1.2.7, which requires description within the synchronized media experience.
Frequently asked questions
What does WCAG 1.2.7 Extended Audio Description require?
It requires that where pauses in the foreground audio are insufficient to allow audio descriptions to convey the sense of the video, extended audio description is provided for all prerecorded video content in synchronized media. Extended audio description works by periodically freezing the video image while a longer narration describes what is happening, then resuming playback. It is a Level AAA success criterion introduced in WCAG 2.0 and unchanged in WCAG 2.2.
How is extended audio description different from standard audio description?
Standard audio description (1.2.5, Level AA) squeezes narration into the natural pauses between dialogue — the description can never be longer than the gaps allow. Extended audio description removes that limit: the video and its audio are paused, the narrator describes the visuals at whatever length is needed, and then the video resumes. The total running time of the described version becomes longer than the original.
When is extended audio description actually needed?
Only when the standard approach cannot work — that is, when the video is so dense with dialogue or continuous narration that the pauses are too short to describe essential visual information. A talking-head lecture where the speaker says everything that appears on screen needs no description at all. A fast-cut demo where actions on screen are never verbalized, and the presenter never stops talking, is exactly the case 1.2.7 targets.
Does 1.2.7 replace 1.2.5 Audio Description?
No, it builds on it. 1.2.5 (Level AA) requires audio description for all prerecorded video in synchronized media. 1.2.7 (Level AAA) adds that when ordinary pauses are insufficient, the description must be extended by pausing the video. If natural pauses are sufficient to convey the sense of the video, standard audio description also satisfies 1.2.7 — the extension is only required where it is needed.
How do I deliver an extended described version technically?
The most common approach is a separately produced alternate version of the video with the freezes and extended narration edited in, offered via a link or player menu next to the original. More dynamic approaches script the pauses in the player: JavaScript pauses the video at cue points, plays a description audio clip (or has a screen reader announce text), then resumes. The SMIL standard also supports this, though player support is limited; a produced alternate file is the most robust route.
Is 1.2.7 required for ADA, Section 508, or EN 301 549 compliance?
Generally no — those standards reference WCAG Level AA, and 1.2.7 is Level AAA. Standard audio description (1.2.5) is the AA requirement. Extended description is an enhancement most valuable for instructional and training video, where missing the visual steps means missing the content, and for organizations that have committed to AAA media accessibility.
Related Success Criteria
Provide alternatives for prerecorded audio-only and video-only content.
Captions are provided for all prerecorded audio content in synchronized media.
Audio description or full text alternative is provided for prerecorded video content.
Captions are provided for all live audio content in synchronized media.
Audio description is provided for all prerecorded video content.