WCAG 1.2.5 Audio Description: How to Make Video Content Accessible

WCAG 1.2.5 Audio Description: what this success criterion requires

WCAG 1.2.5 is the Web Content Accessibility Guidelines success criterion that covers audio description for prerecorded video. It is a Level AA requirement. The goal is simple: people who cannot see the on-screen action must be given the information they need to understand the video.

When WCAG 1.2.5 applies and who it affects

This criterion applies to prerecorded synchronized media that conveys important visual information not otherwise available in the audio track. It affects content owners, e-learning teams, broadcasters who publish archived video, and platform owners who host or embed prerecorded videos.

What qualifies as acceptable audio description

There are three practical ways to meet 1.2.5 for prerecorded video:

• Provide a separate audio-described version of the video where the necessary visual information is narrated during natural pauses.
• Provide an alternative audio track that users can select in the player that contains the description mixed or authoritatively timed to the video.
• Provide an extended transcript that contains time-coded descriptions of the visual information, when a separate audio option is technically infeasible.

Key rules for descriptions

• Include only information needed to understand the content. Do not repeat dialogue unless it clarifies meaning.
• Place descriptions during natural pauses so they do not overlap spoken content.
• Keep descriptions concise and focused on what a sighted viewer sees but an audio-only user would miss.

Step-by-step: add audio description to an existing prerecorded video

1. Watch the video and mark timestamps where visual-only information is essential (e.g., facial expressions, scene changes, on-screen text, actions).
2. Write short description scripts for each marked segment. Aim for 3 to 12 seconds per description depending on available silence.
3. Record the narration with a clear voice and consistent level. Use the same codec and sample rate as the original audio when possible to avoid playback issues.
4. Insert descriptions into the audio track at the chosen timestamps. Create either a single audio-described master file or a separate audio track file.
5. Publish both versions or enable the additional track in your player. Provide a clear label such as “Audio-described version” or “Audio description” in the player UI.
6. Provide an alternative: if you cannot deliver a second audio option, publish an extended, time-coded transcript alongside the video.

Technical options and player support

Choose the option that fits your platform and user needs. Common approaches are:

• Publish two video files: a standard version and an audio-described version. This is straightforward for download and many streaming setups.
• Provide a selectable audio track in the player so users switch audio without reloading the video. Many modern players support multiple audio tracks.
• Use a descriptions track where the player supports the HTML track kind for descriptions. Where support is limited, fall back to a second file.

File formats and media handling

When possible, keep media formats consistent. Use the same container and codecs for both versions of the video. Label metadata so accessibility tools and users can detect the presence of an audio description option.

Testing and verification checklist

Check each item to confirm the description meets WCAG 1.2.5 in practice, not just on paper.

• Can a user select the audio description option in the player without technical barriers?
• Do descriptions appear only during pauses and not over spoken dialogue?
• Does the description add missing visual information needed to understand the story, action, or key content?
• Is the narrator voice clear, at a comfortable volume, and distinguishable from other audio?
• If offering an extended transcript, does it include timestamps and describe visual details in reading order?
• Test with keyboard navigation and screen readers to confirm controls and labels for the audio-described version are accessible.

Common mistakes to avoid

• Waiting until the last minute. Adding descriptions in post without planning often causes overlaps or omissions.
• Describing every visual detail. That overwhelms listeners. Focus on what is necessary to follow the content.
• Using machine-generated audio descriptions as the only option. Auto descriptions can help but usually miss context and nuance.
• Hiding the audio description option behind unclear menus or combined controls that are hard to find for keyboard users.

When an extended transcript is acceptable

An extended, time-coded transcript can be acceptable when producing a second audio track or a single-file description is not feasible. The transcript must include all visual information in a clear sequence and show timestamps so users can locate the described segments in the video. Make the transcript downloadable and present it near the video player.

Practical decision framework for small teams

Use this quick framework to decide what to do for each video:

• If the visuals carry critical information the audio does not, provide an audio-described version or track.
• If visuals add context but do not change comprehension, consider an extended transcript and a clear on-page summary.
• If resources are limited, prioritise content with the largest audience or that is legally required, and schedule batch production for similar videos.

Final checklist before publishing

• Audio description available and labelled in the player.
• Descriptions well timed and concise.
• Alternative extended transcript provided when needed, with timestamps.
• Controls accessible by keyboard and announced by screen readers.
• Metadata or captions indicate the presence of audio description in the page content.

Meeting WCAG 1.2.5 improves access for people who are blind or have low vision and increases the reach and usability of your video content. Use the step-by-step approach and checklist above to turn inaccessible prerecorded videos into compliant, usable media.