WCAG compliance for audio and video

What WCAG 2.2 requires for audio and video

WCAG requires a transcript for audio-only content, synchronized captions for video and audio description for visuals the soundtrack does not explain.

WCAG, the Web Content Accessibility Guidelines, is the international standard for digital accessibility published by the World Wide Web Consortium (W3C). The current version, WCAG 2.2, became a W3C Recommendation in October 2023.

It defines testable success criteria at three conformance levels: A (the minimum), AA (the level most laws and policies target) and AAA (enhanced). Each version builds on the last, so content that conforms to WCAG 2.2 also conforms to 2.1 and 2.0.

Audio and video fall under Guideline 1.2, Time-based Media. Its criteria are unchanged from WCAG 2.0 through 2.2, so the media requirements apply whichever version a law or contract cites. Here is what each criterion requires, per the W3C specification.

Level A: the minimum for any conformance claim

1.2.1 Audio-only and Video-only (Prerecorded). Prerecorded audio-only content, such as a podcast episode, needs a text transcript that conveys equivalent information. Prerecorded video-only content, such as a silent product demo, needs either a text alternative or an audio track describing what happens on screen.

1.2.2 Captions (Prerecorded). All prerecorded video with audio needs synchronized captions: not a transcript beside the player, but captions timed to the speech that carry the dialogue and meaningful sound information. The only exception is media that is itself an alternative for text and is labeled as such.

1.2.3 Audio Description or Media Alternative (Prerecorded). Prerecorded video needs either an audio description track or a full media alternative: a descriptive transcript that captures both the dialogue and the important visual information. At Level A, a thorough descriptive transcript satisfies this criterion on its own.

Level AA: what most laws require

1.2.4 Captions (Live). Live audio in synchronized media, such as a webinar or a live stream, needs captions produced in real time, usually by a live captioner or a CART provider.

1.2.5 Audio Description (Prerecorded). At Level AA the transcript alternative is no longer enough for visual content. Prerecorded video needs an audio description track that narrates important visual information during pauses in the dialogue. If the audio already conveys everything meaningful that happens on screen, no separate track is needed.

Level AAA: enhanced accessibility

Level AAA adds sign language interpretation (1.2.6), extended audio description where dialogue leaves no room for narration (1.2.7), a full media alternative for all prerecorded synchronized media (1.2.8) and live text alternatives for live audio-only content (1.2.9).

Few organizations target AAA across an entire site, but 1.2.8 is worth knowing: a complete descriptive transcript published alongside every video is the gold standard, and it is the one AAA media criterion that automation makes inexpensive.

The laws that turn WCAG into a legal requirement

WCAG itself is a voluntary technical standard; its legal force comes from the laws and regulations that reference it, and that list keeps growing.

The ADA: US businesses and governments

The Americans with Disabilities Act predates the web, but the Department of Justice has long taken the position that it applies to digital content. In April 2024 the DOJ published a rule under ADA Title II requiring state and local government web content and mobile apps to conform to WCAG 2.1 Level AA.

An interim final rule published in April 2026 extended the compliance dates by one year. Public entities serving 50,000 people or more now have until April 26, 2027, and smaller entities and special district governments until April 26, 2028 (source: ADA.gov). For public universities, transit agencies and city governments, captions on prerecorded video are an explicit regulatory requirement, not an interpretation.

For private businesses under Title III, the websites of public accommodations such as retailers, banks, hotels and streaming services, there is no formal technical standard yet. Courts and DOJ settlements consistently point to WCAG as the benchmark, digital accessibility lawsuits run to thousands of filings a year in US courts, and video without captions is among the barriers those complaints cite.

Section 508: federal agencies and the vendors that sell to them

Section 508 of the Rehabilitation Act requires federal agencies to make their information and communication technology accessible. The US Access Board's revised 508 standards, in effect since 2018, incorporate WCAG 2.0 Level AA by reference (source: access-board.gov). That includes the media criteria: captions on prerecorded video and audio description at Level AA.

Section 508 also reaches contractors and vendors selling to the federal government, which is why accessibility conformance reports (ACRs, based on the VPAT template) now show up in procurement. If your training videos or product demos are part of a federal sale, they need captions.

The European Accessibility Act: enforcement is live

The European Accessibility Act (Directive (EU) 2019/882) has applied since June 28, 2025. It covers a broad range of digital products and services sold in the EU, including e-commerce, banking, audiovisual media services, e-books and transport services, and it reaches non-EU companies selling into the EU market.

Its requirements are assessed against the harmonized European standard EN 301 549, which incorporates the WCAG success criteria, including those for time-based media. Enforcement and penalties are set by each member state, so fines vary by country. US companies selling covered services to EU consumers are in scope.

Broadcast and streaming: FCC rules and the CVAA

In the US, the 21st Century Communications and Video Accessibility Act (CVAA) and FCC rules require closed captions on video programming that aired on US television with captions and is later distributed online. If you republish broadcast content on the web, the caption obligation follows it.

This page is general information, not legal advice. For compliance decisions, consult counsel familiar with your jurisdiction and sector.

How transcripts and captions satisfy each requirement

The pattern across every framework is the same: text alternatives for audio, synchronized captions for video and description of meaningful visuals.

Transcripts

A transcript satisfies 1.2.1 for audio-only content outright: publish an accurate transcript with the episode and the criterion is met. A descriptive transcript, one that includes the relevant visual information alongside the dialogue, also satisfies 1.2.3 at Level A and 1.2.8 at Level AAA. Sonix's automated transcription converts audio to text at 99% accuracy on clear recordings, and the editor lets you add visual descriptions before you publish.

Captions

Captions satisfy 1.2.2, and the key word in the criterion is synchronized: caption text must be timed to the audio. Sonix generates automated subtitles with word-level timing and exports SRT, VTT and TTML files that work with YouTube, Vimeo and every major player and editing suite.

For the full production workflow, from splitting and timing to burning captions into the picture, see our automated subtitles and captions pillar and the guide on how to place subtitles in a video.

SDH subtitles

SDH subtitles go a step further than standard captions by adding speaker identification, sound effects and music cues, which is what deaf and hard-of-hearing viewers need to follow the full soundtrack. WCAG's definition of captions includes non-speech audio information, so SDH-style output is the safer target for 1.2.2. Sonix labels speakers automatically and you add the sound cues in the editor; our SDH subtitles guide covers formats, conventions and workflow.

Audio description

Audio description is the one requirement text cannot fully cover at Level AA, and Sonix does not generate audio description narration tracks. What Sonix provides is the fastest path to the two adjacent obligations: descriptive transcripts that satisfy 1.2.3 at Level A, and accurate scripts and timing that make recording a description track with a voice artist or a text-to-speech tool much faster.

If your conformance target is WCAG 2.1 or 2.2 AA and your videos carry meaningful visual-only information, plan audio description production alongside your captioning workflow.

Live captions

Criterion 1.2.4 applies to live streams and webinars. The Sonix workflow on this page is built for prerecorded media, so for live events book a real-time captioning (CART) provider or ask about Sonix real-time transcription for enterprise, then run the recording through Sonix afterwards so the on-demand version meets 1.2.2 with clean, corrected captions.

Compliance comes down to accuracy and volume

Captions only count toward WCAG when they convey the audio accurately, and most teams hold far more media than a human captioning budget covers.

Unreviewed auto-captions routinely fail manual audits because they garble names, terminology and overlapping speech. Human captioning fixes the accuracy but often costs $100 or more per hour of content, which puts a full archive out of reach.

The workflow below handles both: automated transcription takes care of the volume, and a human review pass in the editor delivers the accuracy that an audit expects.

Who accessible media serves

Compliance is the floor, not the point: the audience for transcripts and captions is far larger than the legal definitions suggest.

Deaf and hard-of-hearing users

The World Health Organization estimates that more than 1.5 billion people live with some degree of hearing loss, around 430 million of them with disabling hearing loss, and projects nearly 2.5 billion by 2050. For this audience, captions and transcripts are not a convenience feature. They are the content.

Non-native speakers

Reading along while listening helps comprehension in a second language. Captions in the original language, plus subtitles in a viewer's first language, turn one video into a global asset. The European Accessibility Act makes multilingual access a practical question too: Sonix translates transcripts and subtitles into 54+ languages, so one workflow covers your German product page and your Japanese training library.

Sound-off viewers

A 2019 Verizon Media and Publicis Media survey of US consumers found that 69% watch video with the sound off in public places, and 80% are more likely to watch a whole video when captions are available. Captions are a performance feature as well as an accessibility one.

Search engines and AI systems

Crawlers cannot listen. A published transcript exposes every product name, question and answer in your media to Google and to the AI assistants that increasingly summarize the web, so a page with a transcript can rank for the long-tail phrases spoken in the recording.

Built for teams with compliance deadlines to meet

Sonix pairs speed with the review tools that compliance work requires.

That means 99% accurate automated transcription on clear audio, a fast in-browser editor for the human pass, SRT, VTT and TTML export, automatic speaker labels, translation into 54+ languages and batch upload for clearing archives of hundreds of files. Universities, media companies and enterprises use Sonix to work through caption backlogs that would take months with traditional captioning services.

Security reviews are covered too: Sonix is SOC 2 Type II certified and GDPR compliant, offers HIPAA-compliant plans and encrypts your files with AES-256. The details are on the security page.