How to Test Video Accessibility and Fix Common Issues
Test captions, transcripts, audio description, and player controls together. This practical workflow helps you find common video accessibility issues, fix them, and retest.
Test the video and its player together. Review captions against the audio, check whether a transcript and audio description or another suitable alternative cover the content, and operate the player controls with a keyboard. Automated checkers can help find some issues, but a person must judge whether the media is understandable and whether the controls work.
This guide gives you a repeatable review and repair process. The right requirements depend on the applicable WCAG version, media type, delivery context, and jurisdiction; a checklist alone does not establish legal compliance.
1. Scope the video before testing
Write down what kind of media you are reviewing. This determines which accessibility questions matter.
- Prerecorded or live: live captions and prerecorded captions may have different requirements.
- Audio-only, video-only, or synchronized media: a video with speech and visuals can carry information through both channels.
- Information sources: note spoken words, meaningful sounds, actions, graphics, and on-screen text.
- Delivery context: record the player, page, and version of the video under review.
- Applicable criteria: identify the WCAG version and jurisdiction or policy that applies to your project.
W3C’s media guidance maps different media types to relevant success criteria. Use it to identify the criteria for your specific content rather than applying one generic checklist to every video. W3C WCAG Understanding documents
2. Test captions against the audio
Play the complete video and compare what you hear with what appears in the captions. A caption track that exists but is inaccurate, incomplete, or difficult to read still needs work.
- Check coverage. Captions should include dialogue and meaningful sounds, such as a warning tone or a sound that changes the meaning of a scene.
- Check synchronization. Captions should appear with the corresponding speech or sound and remain visible long enough to read.
- Check wording. Correct transcription errors, spelling, punctuation, and confusing line breaks.
- Identify speakers. Make it clear who is speaking when that is not obvious from the audio or scene.
- Review sound labels. Use consistent labels for relevant sounds and music; do not label every incidental noise.
- Check collisions. Make sure captions do not obscure important on-screen text, faces, or visual details.
- Review pace. Long or fast captions can be hard to follow. Section508.gov notes that speech over 180 words per minute may be too fast for captions; treat this as general guidance, not a universal legal cutoff.
Section 508 caption guidance describes synchronized captions that include dialogue and important sounds, stay visible long enough to read, and follow consistent conventions. Section508.gov: Creating Accessible Videos
3. Check the transcript and other alternatives
A transcript is a separate text alternative; it does not replace captions when a person needs text synchronized with playback. Put the transcript near the video and make it easy to open with a keyboard and assistive technology.
Check that the transcript matches the version being tested and includes the meaningful content a reader needs. For a video, spoken words alone may not communicate essential visual information. Include descriptions of significant action, graphics, or on-screen text when those details are not conveyed in the audio or another alternative.
Section508.gov gives accessible web pages, plain text, and conformant Word documents as example transcript formats. Choose a format that works for your audience and provide a clear link next to the media. Section508.gov transcript and caption guidance
4. Decide whether audio description is needed
Ask: could someone who cannot see the video understand the information that matters using the audio and available alternatives? If meaningful actions, graphics, or text appear only visually, captions will not convey them. Add audio description or another suitable media alternative as needed for the applicable criteria.
Possible approaches include:
- Integrate description into the existing narration: use pauses or revise the script to describe important visual information.
- Produce an alternative described video: provide a version with the necessary visual details spoken aloud.
- Supply a separate description track or timed-text file: use this when the player supports selecting and playing that alternative.
Choose based on what information is missing, production constraints, and the capabilities of the delivery player. W3C describes these approaches and notes that organizations may outsource description production. W3C: Making Audio and Video Media Accessible
5. Test the player controls
Review the actual embedded player, not only the media file. Confirm that people can find, understand, and operate the caption and audio-description controls.
- Start playback and use the keyboard to reach every relevant control.
- Confirm focus is visible and moves in a logical order.
- Turn captions on and off. Confirm the selected captions actually appear and stay synchronized.
- If description is available, select it and confirm it can be heard or otherwise accessed as intended.
- Check that control names communicate their purpose and that the controls remain usable while the video plays.
- Repeat the checks in the browsers and devices your audience is expected to use.
For covered synchronized-media information and communications technology, Revised Section 508 Standard 503.4 addresses caption and audio-description controls. Section508.gov guidance discusses placing those controls at the same menu level as volume or program selection in specified cases. Those federal requirements apply within their scope; do not assume they govern every project. Section508.gov multimedia guidance
6. Combine automated checks with human review
Automated accessibility tools can flag some technical problems, but they cannot decide whether captions accurately represent a scene or whether a visual description preserves the information a viewer needs. W3C says no single tool can determine whether a site meets accessibility standards and that knowledgeable human evaluation is required.
- Use automated checks for detectable page and player issues.
- Manually watch and listen to the video, review the transcript, and operate the player.
- Where practical, include people with disabilities in evaluation.
- Record what was checked, the evidence, and the result so fixes can be retested.
Section508.gov describes automated, manual, and hybrid evaluation methods. Use tools as part of the workflow, not as a pass/fail verdict on the media. W3C: Test and Evaluate · Section508.gov: Test
7. Fix issues and retest the delivered version
Fix the content or player where the issue occurs, then repeat the relevant checks in the delivery environment. A corrected caption file can still be out of sync if the player loads a different version, and an added description track is not useful if viewers cannot select it.
- Correct caption wording, sound coverage, speaker identification, or timing.
- Add or revise the transcript and include essential visual information in an appropriate alternative.
- Integrate audio description, publish a described version, or configure a supported description track.
- Repair player controls and confirm they are keyboard operable and understandable.
- Replay the affected sections and retest the full video when the change could affect synchronization or playback.
- Document the video version, issue, evidence, fix, and retest result.
For a library of videos, prioritize content with the greatest audience impact and track remediation by version. Caption preparation, transcript work, and audio-description production can be handled in-house or outsourced; verify a provider’s capabilities and terms before engaging them.
Common problems and fixes
| Problem | Why it matters | Fix |
|---|---|---|
| No captions, or captions omit important sounds | Viewers who cannot hear the audio miss dialogue or meaningful sound cues. | Add or correct synchronized captions covering dialogue and relevant sounds. |
| Captions lead, lag, or disappear too quickly | The text is hard to associate with the audio or read in time. | Adjust timing and display duration, then replay the affected sections. |
| Unclear speakers or inconsistent sound labels | Readers may not know who is speaking or what a sound means. | Apply consistent speaker and sound identification where context does not make them clear. |
| Transcript is missing or hard to find | People who need a text alternative may not discover it. | Link an accessible transcript alongside the player and verify it matches the current video. |
| Transcript contains speech but omits essential visuals | Important actions, graphics, or text remain unavailable to someone who cannot see them. | Add that information to the transcript or provide audio description or another suitable alternative. |
| Description exists but cannot be selected | The alternative is unavailable through the delivered player. | Check player support and expose an operable selection control, or choose another delivery approach. |
| Automated scan reports no issues, but the video is still inaccessible | A scan cannot judge all media content or playback behavior. | Complete manual listening, viewing, keyboard, and content review; include knowledgeable evaluators. |
Performance, reliability, and maintenance
For a single video, a complete manual review is usually the core task. For a large library, make the work repeatable: maintain a content inventory, track captions and alternatives by video version, assign issues, and schedule retests when media or players change.
- Reliability: check the actual published embed and assets. A local file or preview may not match what the audience receives.
- Retesting: retain evidence and version details so fixes can be checked against the same content.
- Scale: use automated checks to find suitable technical issues across many pages, while reserving human review for meaning, accuracy, and operability.
- Cost: estimate caption correction, transcript preparation, description production, and player remediation separately. Needs vary by media length and complexity; no single price or time estimate applies.
Or skip the browser setup
If you need a screenshot of the page where the video is embedded for a visual review record, ScreenshotNeo is a website screenshot API and MCP server. A screenshot can help document the page and visible player state; it does not replace caption, transcript, audio-description, keyboard, or human accessibility checks.
One GET request returns an image or PDF. See the ScreenshotNeo API documentation for options.
curl -G "https://api.screenshotneo.com/v1/shot" -d access_key=YOUR_API_KEY --data-urlencode url=https://example.com/video-page -o shot.webp
import requests
r = requests.get("https://api.screenshotneo.com/v1/shot", params={"access_key": "YOUR_API_KEY", "url": "https://example.com/video-page"}, timeout=90)
open("shot.webp", "wb").write(r.content)
const q = new URLSearchParams({ access_key: 'YOUR_API_KEY', url: 'https://example.com/video-page' });
const res = await fetch(`https://api.screenshotneo.com/v1/shot?${q}`);
- Cookie banners, popups, and chat widgets are removed before the shot; each removal step can be turned off.
- Bot checks, blank pages, timeouts, failed loads, and cache hits are not billed. Response headers report the page verdict and billing status.
- An MCP server gives AI agents tools to take screenshots, get page information, and capture PDFs.
- 1,000 screenshots a month are free with no card. Paid plans start at $5 for 3,000 screenshots; every feature is on every plan.
Sign up for ScreenshotNeo’s free plan to get 1,000 screenshots a month with no card.
FAQ
Does adding captions mean audio description is unnecessary?
No. Captions communicate audio information. If meaningful visual information is not conveyed through audio, assess whether audio description or another suitable alternative is needed.
Can an accessibility checker certify that a video is accessible?
No single tool can make that determination. Combine automated checks with knowledgeable human evaluation of the media and player.
Is 180 words per minute a legal caption limit?
The cited Section508.gov guidance presents speech above that rate as potentially too fast for captions. It is general guidance, not a universal legal threshold.
Does the federal player-control guidance apply to every video?
No. The cited Revised Section 508 provisions concern covered ICT. Determine which standard and jurisdiction apply to your content.


