Direct answer: In short, this guide covers wCAG Caption Requirements for Level AA Videos: the practical options, the trade-offs, and what to watch out for.
Key takeaways:
- The Four Success Criteria That Matter: Level A is the floor; Level AA includes everything in A plus the AA criteria.
- 1.2.2 Captions (Prerecorded), Level A: What it says: captions are provided for all prerecorded audio content in synchronized media, except when the media is a media alternative for text
- 1.2.4 Captions (Live), Level AA: What it says: captions are provided for all live audio content in synchronized media.
- 1.2.3 and 1.2.5: Audio Description, Levels A and AA: Captions cover hearing; audio description covers vision, narration of important visual information (actions, scene changes, on-screen text) inserted into pauses in the dialogue.
- When in doubt, test with your own audio before committing to a paid plan.
WCAG, the Web Content Accessibility Guidelines, is the international standard that defines what “accessible video” means. If you publish videos on a website and care about accessibility (or about the laws that reference WCAG, like the ADA in the US and the European Accessibility Act), Level AA is the conformance level you’re almost certainly aiming for. It’s the level most regulations and lawsuits point to.
[ez-toc]
The caption-related requirements live in Guideline 1.2: Time-based Media, specifically success criteria 1.2.2 through 1.2.5. The official wording is dense legal-technical prose, so this guide translates each criterion into a plain-English checklist with practical steps for content creators.
Not legal advice. This article explains what WCAG says. Whether WCAG applies to you legally depends on your jurisdiction, sector, and the specific law in question (ADA, EAA, Section 508, etc.). For compliance decisions, consult a qualified accessibility or legal professional.
The Four Success Criteria That Matter
| Criterion | Level | Applies to | One-line meaning |
|---|---|---|---|
| 1.2.2 Captions (Prerecorded) | A | Prerecorded video with audio | All prerecorded video needs captions |
| 1.2.3 Audio Description or Media Alternative | A | Prerecorded video | Provide audio description or a text alternative |
| 1.2.4 Captions (Live) | AA | Live video with audio | Live streams need real-time captions |
| 1.2.5 Audio Description (Prerecorded) | AA | Prerecorded video | Prerecorded video needs audio description |
Level A is the floor; Level AA includes everything in A plus the AA criteria. So a Level AA target means you must satisfy all four rows above. (Level AAA adds sign-language interpretation and extended audio description, beyond what most organizations target.)
1.2.2 Captions (Prerecorded), Level A
What it says: captions are provided for all prerecorded audio content in synchronized media, except when the media is a media alternative for text and is clearly labeled as such.
Plain English: every video you publish that has meaningful audio needs captions. The only exception is video that exists purely as an alternative to text content (rare).
Practical checklist:
- [ ] Caption every prerecorded video with dialogue, narration, or meaningful sound, no exceptions for “short” videos.
- [ ] Captions must be synchronized, timed to the audio, not a separate transcript file dumped below the video (a transcript is a nice extra, not a substitute).
- [ ] Include speaker identification when it’s not visually obvious who’s talking.
- [ ] Include relevant non-speech sounds:
[applause],[door slams],[ominous music]. Deaf viewers need the sound information, not just the words. - [ ] Captions should be accurate, auto-captions you never reviewed don’t reliably meet this. Edit them.
- [ ] Don’t obscure important visuals, keep captions positioned so they don’t cover faces, graphics, or on-screen text.
1.2.4 Captions (Live), Level AA
What it says: captions are provided for all live audio content in synchronized media.
Plain English: this is the criterion that pushes you from Level A to Level AA for live content. If you stream live video with audio, webinars, live events, broadcasts, you need real-time captions.
Practical checklist:
- [ ] Arrange live captioning for streams: a human CART captioner for high-stakes events, or automatic live captioning (built into platforms like YouTube Live, Zoom, or Teams) where acceptable.
- [ ] Publish the archive with captions afterward, the recorded version is prerecorded media, so 1.2.2 applies to it too. Auto-captions from the live stream are a starting point; clean them up before the replay goes public.
- [ ] Test the caption display on your streaming platform before going live, confirm viewers can actually see and toggle them.
Note the asymmetry: at Level A, live captions aren’t required; at Level AA, they are. This single criterion is why many organizations’ “we’ll add captions later” approach fails an AA audit.
1.2.3 and 1.2.5: Audio Description, Levels A and AA
Captions cover hearing; audio description covers vision, narration of important visual information (actions, scene changes, on-screen text) inserted into pauses in the dialogue.
- 1.2.3 (Level A): you may satisfy this with either audio description or a text-based media alternative (e.g., a full descriptive transcript).
- 1.2.5 (Level AA): the alternative option goes away, prerecorded video needs actual audio description.
Practical checklist:
- [ ] For each prerecorded video, ask: would a blind viewer miss important information? If yes, add audio description.
- [ ] Description goes in natural pauses, it must not talk over dialogue.
- [ ] Describe actions, scene changes, and on-screen text that carry meaning (charts, captions within the video, visual jokes).
- [ ] You don’t need to describe every detail, describe what’s needed to understand the content equivalently.
- [ ] Talking-head videos where everything important is spoken may need little or no description, document that judgment.
The Full Level AA Video Checklist
Putting it together, for every video you publish:
Captions
– [ ] Prerecorded videos: synchronized, accurate captions (1.2.2)
– [ ] Live videos: real-time captions during the stream (1.2.4)
– [ ] Replays/archives: cleaned-up captions, not raw auto-caption output (1.2.2)
– [ ] Speaker IDs and meaningful non-speech sounds included
– [ ] Captions user-toggleable (not burned in) wherever the platform allows
Audio description
– [ ] Prerecorded videos: audio description track for visually important content (1.2.5)
– [ ] Description timed to natural pauses, not overlapping dialogue
Player and page
– [ ] The video player itself is keyboard-accessible (play, pause, volume, caption toggle all reachable by keyboard)
– [ ] Captions don’t auto-play with sound in a way that traps users, standard media controls apply
– [ ] If captions are burned in (some social platforms), ensure the text has sufficient contrast against the video behind it
Common Misconceptions
“Auto-captions are enough.” Auto-captions are a starting point, not a finished product. WCAG requires captions to be an equivalent alternative to the audio, unreviewed auto-captions with wrong words, missing speaker IDs, and no sound descriptions frequently fail that bar, especially on technical or multi-speaker content. Generate automatically, then edit.
“A transcript below the video counts as captions.” No. A transcript is a different (also valuable) accommodation, but 1.2.2 explicitly requires synchronized captions. Do both if you can, transcripts help with SEO and skimming, but the transcript doesn’t replace captions.
“Our videos are too short to need captions.” There’s no duration exception in 1.2.2. A 30-second clip with dialogue needs captions just like a 30-minute one.
“WCAG is just a suggestion.” WCAG itself is a technical standard, not a law, but laws reference it. The ADA (US), the European Accessibility Act, Section 508 (US federal), and the AODA (Ontario) all effectively point to WCAG 2.x Level AA. Treat AA as the practical legal target in most jurisdictions.
“Level AAA is better, so let’s do that.” AAA includes criteria (like sign-language interpretation for all prerecorded video) that the W3C itself says can’t be met for all content. Targeting AAA across the board is not recommended; AA is the recognized standard.
Frequently Asked Questions
Which WCAG version should I follow?
WCAG 2.2 is the current W3C recommendation (published 2023). The 1.2.x media criteria are stable across 2.0, 2.1, and 2.2, so this checklist applies regardless, but cite 2.2 in any accessibility statement you publish.
Do I need captions for videos on social media?
If the video is part of your organization’s web content, yes, WCAG doesn’t exempt social platforms. Practically: use each platform’s caption tools (most now offer auto-captioning you should review), and keep a captioned master copy.
What’s the difference between 1.2.2 and 1.2.4?
1.2.2 covers prerecorded video (Level A); 1.2.4 covers live video (Level AA). Same idea, synchronized captions, but live captioning has to happen in real time, which is why it’s the harder, higher-level requirement.
Does a descriptive transcript satisfy 1.2.5 Audio Description?
At Level A (1.2.3), a text alternative can substitute for audio description. At Level AA (1.2.5), no, you need actual audio description timed to the video. A transcript remains a good supplement either way.
Who is responsible, me or the video platform?
You are, for your content. The platform provides the player; you provide captioned, described media. Choosing a player with proper caption support is part of the job, but uploading an uncaptioned video to an accessible player still fails 1.2.2.
How do I check whether my videos pass?
Manual review against the checklist above is the core method: watch each video with captions on (sound off) and with sound on (eyes closed, description on). Automated checkers can flag missing caption tracks but can’t judge caption quality, that part needs human eyes.
Conclusion
WCAG Level AA for video boils down to four criteria: captions on all prerecorded video (1.2.2), captions on live streams (1.2.4), and audio description on prerecorded video (1.2.5, with 1.2.3’s alternative at Level A). Generate captions automatically if you like, then edit them for accuracy, speaker IDs, and sound descriptions; arrange real-time captioning for live events; and add audio description where visuals carry meaning. Do that consistently, and your video library meets the standard most of the world’s accessibility laws point to.







