1.2.1 Audio-only and Video-only (Prerecorded)
- Level A
- Perceivable
- Since WCAG 2.0
Summary
This criterion protects two groups at once. People who are deaf or hard of hearing cannot get the content of a podcast episode, a recorded voice note, or an audio announcement, so prerecorded audio-only content needs a text alternative that carries everything the audio carries. People who are blind or have low vision cannot get the content of a silent product demo, an animated diagram, or a looping instructional clip, so prerecorded video-only content needs either an equivalent text description or an audio track that narrates what happens on screen.
The auditor's first job is classification: grade what the media actually IS. Audio with no meaningful visuals is audio-only. Video with no meaningful audio is video-only. Media with synchronised sound and picture is synchronized media and belongs to SC 1.2.2 Captions (Prerecorded) and its siblings, not here. The second job is fidelity: an "alternative for time-based media" means a full transcript or description that presents equivalent information (dialogue, speakers, meaningful sounds, on-screen actions), not a one-line summary caption. The single exception is media that is itself an alternative for text on the page and is clearly labelled as such.
Official wording
For prerecorded audio-only and prerecorded video-only media, the following are true, except when the audio or video is a media alternative for text and is clearly labeled as such:
Prerecorded Audio-only:
An alternative for time-based media is provided that presents equivalent information for prerecorded audio-only content.
Prerecorded Video-only:
Either an alternative for time-based media or an audio track is provided that presents equivalent information for prerecorded video-only content.
EN 301 549 mapping
- Web pages
- Clause 9.1.2.1
- Software and native apps
- Clause 11.1.2.1
- Closed functionality (kiosks, terminals)
Closed systems must provide visual output for pre-recorded audio-only content (clause 5.1.5).
Closed systems must provide speech output for pre-recorded video-only content (clause 5.1.3.7).
Clause 9.1.2.1 applies this criterion to web pages unchanged and clause 11.1.2.1 applies it to software that supports assistive technologies. On closed systems, pre-recorded audio-only content must gain visual output (clause 5.1.5) and pre-recorded video-only content must gain speech output (clause 5.1.3.7).
Clause references are to EN 301 549 V3.2.1 (2021-03), the harmonised European standard. Descriptions are our own summary, not the text of the standard.
In practice
Web
Sweep the page for audio and video elements, embedded players, and podcast or voice-note widgets. Classify each: play it and check whether it has both tracks. In browser developer tools you can confirm a suspiciously silent video really has no audio track (inspect the element, or check whether the player's volume control is present and functional); a video with an empty or silent audio track is graded as video-only.
For audio-only content, look for a transcript on the page or a clearly labelled transcript link placed near the player. Read it against the audio: it must identify speakers where that matters and include meaningful non-speech sounds, not just the words. A transcript buried three navigation levels away with no link from the media is a practical failure; the alternative must be discoverable from the content it substitutes.
For video-only content, accept either a full text description near the video (describing the actions, states, and any on-screen text in order) or an audio track added to the video that narrates the same information. A caption element saying "Product demo" under a two-minute silent walkthrough is not equivalent information. Finally, check the exception: if the video merely re-presents adjacent text (for example, an animation of steps that are written out in full next to it) and is labelled as an alternative to that text, it passes via the exception.
iOS
The European standard applies this requirement to software as well as web content, so native screens are graded the same way. The classic iOS failure spots are onboarding animation sequences, video splash tutorials that autoplay on first launch, and voice-message features in messaging or support screens.
Walk the app's first-run flow and any help or tutorial sections, since that is where silent demonstration videos concentrate. For each clip, check whether it is genuinely video-only (no narration) and whether equivalent text exists on the same screen or one obvious tap away: a "Read the steps" link, an expandable text version, or full step text rendered alongside. For voice notes and other recorded audio the app plays back, check whether the app offers a transcript view (many messaging patterns now include one); if the audio is content the app presents, an alternative must exist somewhere discoverable in the app, not only on a marketing website. With VoiceOver running, confirm the transcript or description is actually reachable and readable, since a visually present transcript inside an inaccessible custom view helps nobody.
Note for closed systems: where software runs on hardware the user cannot extend with assistive technology (kiosks and similar), the European standard goes further, requiring audio-only content to also be available as visual output and video-only content to also be available as speech output, because users cannot bring their own tools.
Android
The same grading applies. Hunt in the usual places: onboarding carousels with looping animation clips, in-app tutorial videos, feature-tour screens after an update, and voice-message recording and playback features.
For each piece of media, classify it first. An animated vector illustration that is purely decorative is not "media" carrying information and is handled under 1.1.1 rather than here; a silent screen-recording that teaches the user how to complete a task is video-only content and needs a text description or an audio track. Check where the alternative lives: text on the same screen, a clearly labelled in-app link, or a transcript panel in the player itself all work. Verify with TalkBack that the alternative is exposed to assistive technology. For audio playback features, check for a transcript affordance in the playback UI; if the app presents recorded audio as content and offers no text equivalent anywhere in the app, grade a fail on the screens where the audio is presented.
As on iOS, on closed-system deployments (for example an Android-based kiosk build), the European standard additionally requires visual presentation of audio-only content and spoken presentation of video-only content.
Pass and fail examples
Passes:
- A podcast episode page with a full transcript directly below the player, identifying each speaker and noting meaningful sounds.
- A silent animated GIF-style clip demonstrating a three-step gesture, with the same three steps written out in text immediately beside it.
- A video-only assembly demonstration that ships with an optional narration audio track describing every action shown.
- A silent animation that simply illustrates the instructions already written in full on the page, clearly labelled as an illustration of those steps: the media-alternative-for-text exception applies.
- A voice-notes feature that generates and displays a text transcript of each recorded message in the conversation view.
Fails:
- An embedded audio interview with no transcript anywhere, or with only a two-sentence editorial summary of a forty-minute conversation.
- A silent onboarding video walking through account setup, with no text version of the steps and no audio track.
- A "transcript" that is actually show notes: topic bullets and links, omitting most of what was said.
- A silent product demo whose only text is the caption "Watch how it works".
- A transcript that exists but is rendered inside a custom view no screen reader can read, or is only published on an external site with no link from the media.
Not a fail under this criterion:
- A decorative looping background video with no informational content: that is decoration, handled under SC 1.1.1 Non-text Content (marked decorative), and a transcript would be pointless.
- A video with meaningful narration and meaningful visuals lacking captions: that is synchronized media, graded under SC 1.2.2 Captions (Prerecorded).
- A narrated video whose important visual details are not spoken: an audio description gap, graded under SC 1.2.5 Audio Description (Prerecorded).
- A live audio stream with no transcript: live content is out of scope here (SC 1.2.9 Audio-only (Live) covers it at Level AAA).
Commonly confused with
- SC 1.2.2 Captions (Prerecorded). Captions apply to synchronized media, where audio and video carry information together. If the media has both tracks, file caption findings under 1.2.2; if it has only one, you are here. Misclassifying a silent video as "missing captions" is the most common misfiling in this group.
- SC 1.2.5 Audio Description (Prerecorded). Audio description also targets synchronized media: it adds narration of visual detail to a video that already has an audio track. A video-only clip that gains a narration track satisfies 1.2.1 directly; 1.2.5 is about videos where the existing soundtrack leaves visual information unspoken.
- SC 1.1.1 Non-text Content. Decorative or trivially simple media belongs there. A decorative looping background video with no information needs 1.1.1 treatment (marked decorative), not a transcript. Grade under 1.2.1 only when the media conveys information a transcript or description could carry.
- SC 1.2.3 Audio Description or Media Alternative (Prerecorded). The similarly worded Level A sibling for synchronized media. The split is by media type: single-track media here, dual-track media under 1.2.3 and its Level AA siblings.
- Transcript quality versus presence. A transcript that exists but omits substantial content fails this criterion; do not downgrade an inadequate alternative to a best-practice note. Equivalent information is the normative bar.
How AUDITSU tests this
AUDITSU's audit walkthrough includes prerecorded media checks on every screen you review. Where a screen contains audio or video, the walkthrough first asks you to classify it (audio-only, video-only, or synchronized) so the finding lands under the right criterion, then asks the 1.2.1 questions for single-track media: is there a transcript or equivalent text description, is it discoverable from the media, and does it carry equivalent information rather than a summary? With the screen reader running, you also confirm the alternative is actually readable. Each question records a pass, fail, or not applicable result per screen with evidence attached, so a missing podcast transcript and a silent tutorial video are reported against the exact screens where they appear.
For the full guided workflow, see the audit platform.