Video Accessibility Checklist: What to Check Before Publishing

Video Accessibility Checklist: what to check before publishing

This Video Accessibility Checklist covers the items to verify before you publish any recording. Read the quick checklist first, then follow the technical steps and simple tests to make sure captions, audio description, transcripts and the player itself work for everyone.

Quick pre-publish checklist

• Captions for prerecorded audio present and edited for accuracy
• Live captioning where required for events
• Text transcript available
• Audio description or extended description when visual information is essential
• Accessible media player with keyboard and screen reader support
• Metadata and language declared correctly
• Final test on desktop, mobile and with assistive technology

Required accessibility elements

Captions (for deaf and hard-of-hearing viewers)

Provide captions for all spoken words and relevant non-speech sounds. Use a sidecar file such as WebVTT (.vtt) or SRT (.srt) rather than burning captions into the picture. Edit any automated captions before publishing. Ensure speaker changes are clear when multiple people talk.

Live captions

For live streams or events, use a trained captioner or a high-quality live captioning service. If you rely on automatic live captions, include a note that captions are auto-generated and may contain errors. Live captioning is covered by WCAG success criteria for live media captions.

Transcripts

Provide a full text transcript for every video. The transcript should include spoken words, key non-speech sounds, and scene descriptions where those are needed to understand the content. Transcripts also help people who use screen readers and provide a text version for search indexing.

Audio description

Add audio description or an extended description when visual details are essential for understanding the content. Script the description so it fits natural pauses. Offer the description as a separate audio track or as an alternate version of the video.

Sign language interpretation

Offer sign language interpretation when required by policy or when it significantly improves access. If you include an interpreter on screen, ensure the framing, contrast and size are sufficient for comprehension and the interpreter track is not cropped on small screens.

Language and metadata

Declare the video language and the caption language in the player or file metadata. Use accurate titles, descriptions and chapter markers so assistive tools and users can find relevant content quickly.

Standards and legal context

Design to WCAG 2.1 success criteria that cover timed media. Common references include captions for prerecorded and live media and audio description requirements at levels A and AA. In the UK, consider the Equality Act 2010 and the Public Sector Bodies (Websites and Mobile Applications) Accessibility Regulations 2018 when publishing public sector or regulated content.

Technical specifications and file formats

Caption file formats

• WebVTT (.vtt) is the recommended format for HTML5 video tracks
• SRT (.srt) remains widely used and is compatible with many captioning tools
• TTML/DFXP is used in broadcast and some enterprise systems

Sidecar vs burned-in captions

Use sidecar captions when possible. They can be toggled off, styled by the user and read by assistive tools. Burned-in captions are acceptable when you must ensure captions are always visible, but they cannot be resized or turned off.

Styling and timing

Keep captions readable. Prefer a maximum of two lines on screen. Keep each caption long enough to read. Avoid overlapping captions and make sure caption timing matches the audio. Label speakers when necessary.

Accessible player features to verify

• All controls reachable with keyboard only and operable with standard keys
• Visible focus indicator when tabbing through controls
• Controls have accessible names for screen readers, such as play, pause, captions and volume
• Caption tracks exposed via the HTML5 <track> element or the player API
• Screen-reader friendly announcements for state changes, for example when captions are turned on
• Fullscreen and picture-in-picture accessible from keyboard and announced to assistive tech

Authoring workflow

1. Create a verbatim script or use the published script if available.
2. Generate captions from the script or from a transcription service.
3. Edit captions for timing, speaker ID and sound cues.
4. Produce a plain text transcript that includes non-speech sounds and scene notes.
5. Write audio description script and record the track or create an alternate version.
6. Attach sidecar caption files and metadata to the final package.
7. Test on target devices and with assistive technologies.

Testing and quality assurance

Run these checks before you publish. Test on desktop and mobile. Test with keyboard only. Use at least one screen reader such as NVDA, VoiceOver or TalkBack and verify the player announces changes and exposes captions. Play the captions and read along to confirm accuracy and timing. Confirm the transcript matches the final edit. Check the audio-description track plays when selected.

Quick manual tests

• Turn captions on and off. Verify they match and are readable.
• Navigate all controls with the keyboard and note any traps.
• Open the transcript in a browser or text reader and scan for missing content.
• Play the video with the screen reader active and verify labels and state changes.
• Resize the browser and mobile screen to ensure captions and interpreter windows do not get cropped.

Common mistakes and how to fix them

• Relying on unedited auto-captions. Fix by editing or using a human captioner for accuracy.
• Burned-in captions that are too small on mobile. Use sidecar captions or increase font size.
• Missing non-speech descriptions. Add sound cues such as [applause] or [door closes] in captions and transcripts.
• Player controls hidden from keyboard. Ensure controls receive focus and have visible focus styles.
• No transcript. Create a transcript and link it from the video page and metadata.

Final pre-publish checklist (12 quick checks)

1. Captions file attached and edited for accuracy
2. Caption toggle visible and labelled in the player
3. Transcript published and linked near the video
4. Audio description track available or an alternate described version ready
5. Language and caption language declared in metadata
6. Player operable with keyboard and announces state changes
7. Caption timing and speaker ID verified
8. Interpreter visible and properly framed, if used
9. Files use web-friendly caption format (for example .vtt) and encoding
10. Tested with a screen reader and on at least one mobile device
11. Legal or organisational requirements checked (for example public sector rules)
12. Accessibility statement updated if required

Follow these checks and you reduce the risk of publishing inaccessible video. If you have a long catalogue, prioritise high-use content for immediate remediation and add accessibility into your production workflow.