thearchivist.fun
Glossary · Evergreen reference

Captions

Captions let a viewer follow spoken words and meaningful sounds as a video or other synchronized media plays. Useful captions can identify speakers and communicate audio information such as laughter or an alarm when it matters to understanding. Automatic output often needs review for names, slang, overlap, and timing. In research, keep captions distinct from text visibly added to the picture by an uploader. Both can make claims, but they come from different layers of the artifact. If you quote a caption, verify whether it matches the audible speech and identify the version you inspected. Captions do not necessarily convey all important visual information. A descriptive transcript or other visual description may be needed for a reader who cannot see the action, chart, or on-screen text.

Context and provenance

Time-aligned text that communicates speech and relevant audio in media.

Sources and evidence

  1. W3C WAI: Captions and subtitles

Editorial review: 2026-09-23. New evidence may change this account.

Report a problem with this record