1.2.5 Audio Description (Prerecorded)
- Level AA
- Perceivable
- Since WCAG 2.0
Summary
Audio description exists for people who are blind or have low vision and cannot see what happens on screen. A narrator uses the natural gaps in the existing dialogue and soundtrack to voice the important visual-only information: who entered the room, what the on-screen text says, what the chart shows, which button the presenter just clicked. Without it, a video's meaning can silently depend on pictures the listener never receives.
The auditor's mental model has two steps. First, watch the video with the screen covered (or eyes closed) and note every point where you lose information that a sighted viewer gets. Second, check whether an audio described version or track restores that information. At Level AA the transcript escape hatch from Level A is gone: a text alternative no longer satisfies this criterion, the description must be audio. The flip side is just as important: if the soundtrack already carries all the information (a presenter who says everything they show, a talking-head interview), there is nothing left to describe and the video passes with no AD at all. Failing a talking-head video for "missing audio description" is the most common false fail under this criterion.
Official wording
Audio description is provided for all prerecorded video content in synchronized media.
EN 301 549 mapping
- Web pages
- Clause 9.1.2.5
- Software and native apps
- Clause 11.1.2.5
Clauses 9.1.2.5 (web) and 11.1.2.5 (software) apply this criterion unchanged, so the same requirement binds web pages and native app user interfaces.
Clause references are to EN 301 549 V3.2.1 (2021-03), the harmonised European standard. Descriptions are our own summary, not the text of the standard.
In practice
Web
Inventory every piece of prerecorded video with a soundtrack (synchronized media): product demos, promotional films, tutorials, webinar recordings, embedded third-party players. For each one, play it through once without looking at the screen and list the visual-only information: on-screen text and captions burned into the picture, demonstrations, scene changes, speaker identification where voices are ambiguous, data visualisations.
Then look for the described option. On the web it usually takes one of two forms: a separate audio track selectable in the player (an alternate track menu, often labelled "Audio description" or "Described"), or a second, clearly linked version of the video with description mixed in ("Watch the audio described version"). Both satisfy the criterion; check that the described option is discoverable from the original video, actually plays, and that the narration covers the gaps you listed. In devtools you can confirm a <video> element carries multiple audio tracks, but the practical test is playing the described option and listening.
Grade "no description needed" honestly but strictly. A screen recording where the presenter narrates every action passes; the same recording where the presenter says "just click here and fill this in" while the camera does the work does not, because the dialogue no longer carries the visual information.
iOS
The criterion applies unchanged to software under the European standard, so native apps are graded exactly as the web is: prerecorded video with sound in the app needs audio description unless the soundtrack already carries all the information.
On iOS, described audio ships as an alternate audio track in the media item. AVFoundation exposes these through media selection: a well-built player offers an audio track picker, and tracks correctly tagged as describing video are picked up automatically when the user enables Settings, Accessibility, Audio Descriptions. Test both routes on device: enable the system Audio Descriptions setting, play the video, and listen for the described track; then check the player's own track selector for a description option. If the app uses a custom player, confirm the track picker is actually reachable (and reachable with VoiceOver running, though the name and role of the picker itself are graded under other criteria).
Common iOS failure spots: onboarding and feature-tour videos with burned-in text and no narration, marketing videos dropped into the app with only the original soundtrack, and apps that ship a described track but never surface a way to select it and ignore the system setting.
Android
The same rule applies to Android apps under the European standard: prerecorded synchronized media in the app needs audio description where the soundtrack does not already carry the visual information.
Described audio is delivered as an alternate audio track. Media3 (ExoPlayer) supports track selection and can honour role flags that mark a track as describing video, and Android's system-level Audio description toggle in Accessibility settings signals that preference to apps. Test on device: enable the system Audio description setting, play the video, and listen for the described track; then open the player's track or settings menu and look for a description option. For streaming content, confirm the described track is actually present in the delivered manifest rather than only in the production masters.
Common Android failure spots: help and tutorial videos that demonstrate gestures silently, players built on Media3 where the extra audio track exists but the track selection UI was never wired up, and TV or tablet variants that drop the described track the phone app carries.
Pass and fail examples
Passes:
- A product demo video whose player offers an "Audio description" track; with it enabled, a narrator voices the on-screen steps between the presenter's sentences.
- A promotional film with a prominently linked "audio described version"; the alternate version narrates scene changes and on-screen text.
- A recorded webinar where the presenter consistently reads out every slide heading, figure, and demonstrated action: the dialogue already carries the visual information, so no separate description is required.
- A talking-head interview: two people speaking to camera, no meaningful visual-only information. Nothing to describe, passes as is.
- An iOS app whose tutorial video plays a described audio track automatically when the system Audio Descriptions setting is on.
Fails:
- A silent-demonstration screen recording ("watch how to set up your account") with background music only and no narration or described track.
- A video where key information appears only as burned-in text (prices, statistics, a URL) that no voice ever reads out, and no described alternative exists.
- A described version exists on the production side, but the published player exposes only the original audio track and there is no link to the described version.
- A cooking video where the chef chats about the recipe's history while the technique is shown but never spoken; the dialogue gaps are there and unused, and no AD track is provided.
Not a fail under this criterion:
- A video with a full transcript but no audio description: at Level A that can satisfy SC 1.2.3 Audio Description or Media Alternative (Prerecorded), but at Level AA the transcript route is closed. File the AA finding here; do not also fail 1.2.3 if the transcript is adequate.
- Missing captions on the same video: that is a deaf-side finding under SC 1.2.2 Captions (Prerecorded), not an audio description finding.
- Dialogue gaps too short to fit all the needed description, where the provided AD does its reasonable best: pausing the video to add longer narration is extended audio description, SC 1.2.7, a Level AAA requirement.
- A video-only file with no soundtrack at all (a silent animation): that is not synchronized media, so it is graded under SC 1.2.1 Audio-only and Video-only (Prerecorded), not here.
Commonly confused with
- SC 1.2.3 Audio Description or Media Alternative (Prerecorded). Same videos, different bar. At Level A a full text alternative (a descriptive transcript) is an acceptable substitute for audio description; at Level AA this criterion removes that option and the description must be audible. A video that passes 1.2.3 on the strength of a transcript still fails 1.2.5.
- SC 1.2.2 Captions (Prerecorded). The opposite sensory direction. Captions turn audio into text for people who cannot hear; audio description turns visuals into audio for people who cannot see. One video usually needs both, and they are graded and evidenced separately.
- SC 1.2.7 Extended Audio Description (Prerecorded). The Level AAA sibling for videos whose dialogue gaps are too short to carry the description, allowing the video to be paused while longer narration plays. If standard AD covers what fits in the gaps, do not fail an AA audit for wanting extended description.
- SC 1.2.1 Audio-only and Video-only (Prerecorded). Covers media of a single type: a silent video or a standalone audio file. This criterion covers synchronized media only, video with an accompanying soundtrack. Route silent animations and podcasts to 1.2.1.
How AUDITSU tests this
AUDITSU's audit walkthrough asks a media question on every screen that contains prerecorded video: does the video convey visual-only information, and if so, is an audio described track or version available and adequate? The walkthrough prompts you to play the video without watching it, list what a listener would miss, then locate and play the described option before grading. Each screen records a pass, fail, or not applicable result with evidence attached (the video, the track menu, the described version link), so the "no visual-only information, nothing to describe" pass is documented rather than left as an unexplained N/A.
Because the European standard applies this requirement unchanged to both web content and software, the same walkthrough question covers your website, iOS app, and Android app, with the platform-specific checks (system Audio Descriptions setting, player track selection) built into the per-platform guidance.
For the full guided workflow, see the audit platform.