WCAG Captions Checklist for Video (2026)

Rasif Ali KhanRasif Ali Khan5 min readVideo

A practical WCAG captions checklist for prerecorded video in 2026. What to verify on the file and the player before you call captions done.

On this page

Transcribe faster with File Transcribe

Upload audio or video, get speaker labels, timestamps, and editable text free to try.

Try it free

A WCAG captions checklist for video is a pass/fail you can run on one recording. It is not a certificate that your whole site conforms to WCAG. Captions are one success criterion. Keyboard access, contrast, audio description, and the rest of the page are separate work.

This checklist is for prerecorded video in 2026: lessons, marketing pages, YouTube, and webinar replays. The official criterion to read is WCAG 2.2 understanding captions (prerecorded) (success criterion 1.2.2, Level A). Live captions are 1.2.4 and a different production. Tool shopping: best AI caption tools for accessibility.

The checklist

Copy this into the ticket for the video. Every line should be yes before you publish.

CheckPass looks like
Captions existA timed track, not "we have a blog post of the script"
They match the speechNames, numbers, and steps match what was said
They are synchronizedLines appear with the speech, not a few seconds late
They can be turned on and offSoft SRT/VTT or a real player control, not paint on the picture
Sound cues that matter are included"[applause]" or "[music]" when the sound carries meaning and there is a gap in speech
The player exposes the controlCaptions button is visible and works with keyboard where the page requires it
You watched a sample mutedFirst two minutes plus any demo or price mention

Fail any row and the video is not done.

What this checklist does not cover

  • Audio description for viewers who cannot see the picture (WCAG 1.2.3 and 1.2.5)
  • A transcript-only page as a substitute for captions on video that has important visuals
  • Live events
  • A legal opinion about your jurisdiction

If counsel or a campus accessibility office owns the policy, follow their standard. This page is the production check.

Course-sized version of the same idea: how to make course videos accessible with captions.

Soft captions vs burned-in text

Toggle

Soft SRT/VTT
Yes, if the player supports it
Burned-in text
No

Fix a typo

Soft SRT/VTT
Replace the caption file
Burned-in text
Re-export the video

WCAG-style captions

Soft SRT/VTT
This is the usual pass
Burned-in text
Usually a fail for "user can turn captions on"

Social cuts can still use burned-in text so the words survive reposts. Keep a soft file as the accessible master for your site and YouTube.

How to produce the file this checklist expects

  1. Export the final picture and the final audio. Do not caption a rough cut you will still trim.
  2. Transcribe the master. On File Transcribe, upload the file, edit in the segment editor, export SRT or VTT (subtitle export is on Pro, /pricing).
  3. Fix proper nouns, product names, and any step a learner must follow.
  4. Attach the file in YouTube Studio, your LMS, or the HTML5 track.
  5. Watch muted. Then watch with sound and captions together and look for drift.

Line length and timing fixes: how to fix subtitles for accuracy. YouTube attach steps: how to add captions to YouTube videos.

Speaker labels, music, and silence

Captions are for people who cannot hear the audio. They are not a transcript dumped on screen.

  • Add a speaker name when two people talk and the viewer cannot tell them apart from the picture.
  • Skip a speaker name on every line of a single talking-head video. It wastes the line.
  • Mark meaningful non-speech: laughter that is the joke, a siren, a music sting that replaces dialogue.
  • Do not caption background music word-for-word unless the lyrics are the content.

Speaker cleanup on the transcript, before you export captions: how to add speaker labels to transcripts.

Player checks people skip

The file can be perfect and still fail the viewer.

  • The captions button is not hidden in a menu nobody finds.
  • Turning captions on does not reload the page and lose the timestamp.
  • On mobile, the button is reachable.
  • The default language track is the language of the speech.
  • If you offer a second language, that track is labeled as a translation, not as captions of the original audio. Subtitles for translation and captions for the spoken language are different jobs.

If you only control the file and someone else controls the player, write that down. Your checklist pass is "file ready," not "page conformant."

A 10-minute sample is enough to find the failure

Do not caption a 40-video library and then discover the LMS ignores VTT. Run one video end to end: upload, edit, export, attach, muted watch. Time the edit. Then decide SRT vs VTT based on what the player accepted.

Library-scale version: how to caption an entire course library.

FAQ

Does WCAG require captions on all video?

For prerecorded video that is not a media alternative for text, WCAG 2.2 success criterion 1.2.2 (Level A) requires captions. Read the official understanding doc before you treat a blog post as the spec. Live video is a different criterion.

Is an AI transcript the same as captions?

No. A transcript is text. Captions are timed text synchronized to the media, available in the player. Export SRT or VTT, then attach it.

Can auto captions pass this checklist?

Only after a person checks sync, names, and meaning. Auto captions are a draft. Publishing them untouched fails the "they match the speech" row on any video with jargon.

Do I need audio description too?

If important information is visual only, yes, that is a separate requirement. Captions do not describe a chart that nobody speaks about.

Will File Transcribe make my site WCAG conformant?

No. It produces an editable transcript and SRT/VTT you can attach. Conformance depends on the captions being correct and on the rest of the page and player.

Run the checklist on one video today

Upload the master on File Transcribe, export SRT or VTT, attach it, and watch the first two minutes with the sound off. If you get lost, the captions are not done.

Further reading

Written by

Rasif Ali Khan

Rasif Ali Khan

Founder, File Transcribe

I made File Transcribe to turn recordings into editable text without extra steps. I write these guides from the workflows I use myself, like meetings, podcasts, lectures, and the rest.

All posts →