WCAG 1.2.5 Audio Description (Prerecorded)

What it asks

This tightens 1.2.3 Audio Description or Media Alternative (Prerecorded). At Level A a site could choose between audio description and a full text alternative. At Level AA the choice goes: prerecorded video in synchronised media needs audio description.

Audio description is narration that describes the important visual content, fitted into the pauses in the dialogue. It can be mixed into the main soundtrack, offered as a separate track the viewer selects, or provided as a separate described version of the video. Where the pauses are too short, the Level AAA criterion 1.2.7 allows the video to be paused for extended description, but that is not required here.

As with 1.2.3, a video whose visual content is already carried by the soundtrack needs no extra description. A narrator who says what they are doing as they do it has described the video. The requirement bites on videos where the picture carries information the sound does not.

This is one of the criteria most often marked Does Not Support in real conformance reports, and the honest thing is to say so rather than to stretch a transcript to cover it.

Its official name is success criterion 1.2.5 Audio Description (Prerecorded), at Level AA. It was added in WCAG 2.2, under guideline 1.2 Time-based Media and the principle Perceivable. The W3C’s normative text says, in part:

“Audio description is provided for all prerecorded video content in synchronized media.”

From WCAG 2.2, W3C. Full explanation: Understanding 1.2.5 Audio Description (Prerecorded).

Who it helps

People who are blind or have low vision. Without description, a video that shows a process, a chart, a demonstration or on-screen text leaves them with half the content. People with some cognitive disabilities also benefit from hearing the visual content named.

What passes and what fails

These pass

• A tutorial whose narrator names each control and action as it happens on screen.
• A video with a selectable described audio track in the player.
• A separate described version of each video, linked beside the original.
• A recorded talk where the speaker reads out what is on each slide.

These fail

• Relying on the text alternative that met 1.2.3 to claim Level AA.
• A described version produced for the flagship video only, with the rest of the library undescribed.
• A product tour where features are shown with music and captions on screen, and nothing spoken.
• A recorded presentation where the speaker says "as you can see here" and never says what is there.

How to test it by hand

1. Find every recorded video with sound on the page.
2. Watch each one without looking at the screen. Note every point where something shown would be needed to follow it and was not spoken.
3. If there were such points, look for a described audio track in the player, or a described version of the video.
4. Listen to the described version and check that the description covers those points.
5. Do not accept a transcript or text alternative at this level. Note it as meeting 1.2.3 only.
6. Note each video with important visual content and no audio description.

What automated testing can decide

What a machine can decide

Automated testing can find each video and see whether a description track is declared in the player, which is positive evidence of a described version. Where a page has no video at all, having looked supports a Not Applicable result.

What it cannot

A declared track is the only kind of description automated testing can see. Description mixed into the main soundtrack, which is the most common kind, leaves no trace in the markup. Automated testing also cannot tell whether a video has visual content that needs describing. So the absence of a track is never evidence of a failure, and the presence of one says nothing about whether it describes enough.

How Dreven handles it

Dreven finds the recorded videos on your pages and shows you each one with any audio description found for it. A reviewer watches the video and decides each result, and the report says how it was decided.

The report keeps 1.2.3 and 1.2.5 apart: a site that meets Level A through a text alternative is not credited with Level AA on the strength of it.

No automated check makes a site conform. In Dreven, a person reviews the results and signs the report.

Related criteria

• 1.2.3 Audio Description or Media Alternative (Prerecorded) (Level A). The Level A version, which also accepts a full text alternative.
• 1.2.2 Captions (Prerecorded) (Level A). The same video needs captions for what is said.
• 1.2.1 Audio-only and Video-only (Prerecorded) (Level A). Silent video needs its own alternative.

Questions people ask

All WCAG 2.2 criteria, in plain words