Video descriptions, captions, and audio description
Last updated: September 8, 2026 · 1 min read
Provide equivalents for the video’s audio and visual information.
Captions convey speech and meaningful sounds. Description conveys important visuals missing from the audio. They serve different needs.
Follow these steps
- 1
Name the video
PropertiesAccessibility
In accessible authoring, enter a meaningful video name and indicate whether its audio carries meaning. Starting muted does not mean the source has no meaningful audio.
- 2
Provide a transcript
PropertiesAccessibility
Write the speech, meaningful sounds, and necessary visuals in the transcript/equivalent description field. For a silent video, explain what the viewer needs to understand from the visuals.
- 3
Add captions
PropertiesCaptions
For meaningful audio, enter a language code and reader-facing label, then upload a VTT caption file. The first track becomes default; you can choose another default or remove an incorrect track.
- 4
Describe visual information
PropertiesAccessibility
Indicate whether important visual information is absent from the audio. Use the available timed VTT/SRT, manual timed text, description audio, or described-version link fields. Follow the input example; attaching a file alone does not establish that its content is sufficient. For manual entry, put a time and description on each line, such as “0:05 The box opens,” then choose Add to create the timed description. Set the language code and label for each language.
- 5
Review tracks against the video
EditorPreview
Compare caption language/timing, descriptions, and links with the actual clip. Address Preflight findings and read the equivalents in the accessible version.
Check your result
Check the result in preview, then verify the reader experience in the published version.