F8Documented failure WCAG 1.2.2 Level A Partly scanner-detectable

Do not publish captions that skip dialogue

F8 is a documented failure under the prerecorded captions rule, and it is the caption problem no checker will ever find. The track exists. It is incomplete. Two shapes count. Captions that carry the dialogue and leave out the sounds that matter, and captions that go silent through stretches where people are still talking. What does not count, and this is the part worth telling every video owner, is simplification. Tightening spoken text so a viewer can read it at a comfortable speed is standard practice and does not invalidate a caption. Dialogue has to arrive, word for word or in essence, and the important sounds have to arrive with it. Speaker labeling is our own standard on top of that, because a two-person argument with no names attached is unreadable.

How we find it in an audit

A reviewer watches the video twice, once with sound and once muted with captions on, and writes down the difference. That is the method and there is no shortcut. Dialogue gets checked line by line. Sounds get checked against a list made on the first pass, the knock, the alarm, the laughter, the door. A scanner can tell you a track is missing. It has nothing to say about the one you have.

How affected users experience it

Deaf and hard of hearing viewers are the audience, and for them the caption track is the film rather than an accessory to it. Captions that go quiet for ninety seconds are ninety seconds of story gone, and there is no way to tell from the screen whether nobody spoke or nobody captioned it. Sound effects carry more than people expect. A door slamming ends an argument. An alarm changes what the next line means. Leave those out and the dialogue is all still there, but the scene is not.

Passes vs. fails

Passes: reviewed captions, in sync. Fails: speech with no captions.

Passes

WEBVTT

00:12.000 --> 00:15.000
MAYA: We need to talk.

00:15.500 --> 00:17.000
JONAS: Not now.

00:17.200 --> 00:18.000
[door slams]

Fails

WEBVTT

00:12.000 --> 00:15.000
We need to talk.

<!-- Ninety seconds of argument and a door slam follow, uncaptioned. -->

How this gets tested

The W3C publishes test rules that define what a checker looks for here.

Video element auditory content has accessible alternativeA tool finds candidates, you decide

The other techniques filed under this rule

8 guides on this site are filed under 1.2.2 Captions (Prerecorded). W3C documents this one as a failure of that rule, so it describes a way the rule gets broken rather than a way to pass it.

This guide is our interpretation of W3C technique F8: Failure of Success Criterion 1.2.2 due to captions omitting some dialogue or important sound effects. W3C publishes its techniques as guidance rather than as the standard, and says so on every one of them. The success criterion is what conformance is measured against, and a technique is one documented way to meet it.