Accessible captions: what good looks like
Captions make your video usable for deaf and hard-of-hearing viewers — and for everyone watching muted. Accessibility guidelines are specific about what makes a caption good.
Under the Web Content Accessibility Guidelines (WCAG), prerecorded video with audio needs captions (Success Criterion 1.2.2, Level A). Good captions are accurate, synchronised with the speech, complete, and readable against the video.
What the guidelines ask for
WCAG 2.2 Success Criterion 1.2.2, Captions (Prerecorded), requires captions for all prerecorded audio content in synchronised media, unless the media is itself an alternative for text. The W3C's guidance on captions adds what makes them useful: the words that are spoken, who is speaking when it is not obvious, and meaningful sounds, all in sync with the video.
A checklist for short videos
Accurate
Check names, numbers and slang — automatic transcription gets most words right but not all. Fix them before exporting.
In sync
Captions should appear as the words are spoken. Subtightl times the captions from the speech itself, so they appear as the words are said.
Readable
Strong contrast against the picture, a size you can read on a phone, and not too many words at once. Styles with an outline, shadow or box behind the word stay legible on busy backgrounds.
Out of the way
Keep captions clear of faces and of the platform's own buttons and text at the bottom and right of the screen — move them up if needed.
Complete
Caption the whole clip, not just the hook. If a sound matters to the meaning, say so.
Where burned-in captions fit
Captions burned into the video are always on, which helps muted viewers and people who never find the captions button; a separate caption track lets viewers resize text or switch it off. Short social videos usually get burned-in captions — see open vs closed captions.
- W3C, Understanding WCAG 2.2 — Success Criterion 1.2.2 Captions (Prerecorded) — w3.org
- W3C Web Accessibility Initiative, "Captions/Subtitles" — w3.org/WAI/media/av/captions