Accessible video is video content, together with its player and delivery, that is designed so people with hearing, visual, cognitive, motor or photosensitive disabilities can perceive, understand and operate it. In video accessibility, the term covers the full package: captions, audio description, transcripts, an accessible player and safe visual design, not one single feature.
Accessible video is shaped by recognized standards and laws, most prominently the Web Content Accessibility Guidelines (WCAG), the European Accessibility Act, Section 508 in United States federal procurement and broadcasting rules that require captioning. Beyond compliance, it also helps viewers who watch without sound, in a second language or in noisy places.
Building blocks of accessible video
Accessible video combines several alternative and supplementary tracks, each serving a different group of viewers. No single element meets every need, which is why planning starts with the full list.
| Component | What it provides | Primary beneficiaries |
|---|---|---|
| Closed captions | Dialogue plus sound cues, can be switched on and off | Deaf and hard-of-hearing viewers |
| Open captions | Captions burned into the picture | Viewers on platforms without caption support |
| Subtitles | Translated dialogue, usually without sound cues | Viewers who do not speak the language |
| Audio description | Narration of essential visual information | Blind and low-vision viewers |
| Transcript | Text version of speech and important visuals | Deafblind users, quick readers, search |
| Sign language interpretation | Interpreter shown in or beside the video | Deaf viewers whose first language is signed |
| Accessible player | Keyboard, screen reader and focus support | Motor-impaired and blind users |
The difference between captions and subtitles is often blurred in everyday speech. Captions assume the viewer cannot hear the audio and therefore include speaker identification and non-speech sounds, such as music or a door slamming, while subtitles assume the viewer hears but does not understand the language.
WCAG requirements for video
WCAG 2.1 and 2.2 define testable success criteria for prerecorded and live media. Most laws and procurement rules reference level AA, which is the practical target for most publishers.
| Criterion | Level | Requirement |
|---|---|---|
| 1.2.1 Audio-only and video-only | A | Transcript or alternative for media without both tracks |
| 1.2.2 Captions, prerecorded | A | Captions for all prerecorded audio in video |
| 1.2.3 Audio description or media alternative | A | Description or full text alternative |
| 1.2.4 Captions, live | AA | Captions for live audio content |
| 1.2.5 Audio description, prerecorded | AA | Audio description for prerecorded video |
| 1.2.6 Sign language | AAA | Sign interpretation for prerecorded audio |
| 2.3.1 Three flashes or below threshold | A | No more than three flashes per second |
Player controls are covered by general criteria. Every control must be operable by keyboard, have an accessible name, show visible focus and avoid auto-playing audio for more than three seconds without a way to pause or stop it.
Caption and description quality
Providing captions is not enough; their quality determines whether they help. Broadcast and professional guidelines converge on four properties: accuracy, synchronization, completeness and readable placement.
- Accuracy close to verbatim, with correct names and terms; raw automatic speech recognition often falls short on these.
- Synchronization so that each caption appears with the speech it represents.
- Reading speed typically kept to roughly 15-20 characters per second, depending on audience and guideline.
- Line length of about 32-42 characters and at most two lines per caption.
- Placement that avoids covering faces, lower thirds and on-screen text.
- Sound cues in brackets, such as [phone ringing], when they matter for understanding.
Audio description has its own craft. Standard description fits narration into natural pauses in dialogue, while extended description pauses the video when pauses are too short, which is acceptable online but not in broadcast. The best results come from planning pauses during scripting, a practice often called integrated or inclusive description.
Delivery formats and player support
Accessible tracks must survive the delivery pipeline and be exposed by the player. The format choice depends on the platform and on whether captions should remain switchable.
| Format or method | Typical context | Notes |
|---|---|---|
| WebVTT | Web players and HLS streams | Supports positioning and basic styling |
| SRT | Editing, uploads to platforms | Simple, minimal styling |
| TTML and IMSC | Broadcast, OTT and DASH delivery | Rich styling and positioning |
| CEA-608 and CEA-708 | North American broadcast and streams | Embedded in the video signal |
| Alternate audio track | Audio description in streaming | Selectable in HLS or DASH manifests |
| Burned-in text | Social media autoplay | Always visible, cannot be restyled |
A frequent failure is a player that hides the track. Captions may exist in the manifest while the player offers no button for them, or the control cannot be reached by keyboard. Testing with a screen reader and keyboard only reveals such gaps quickly.
Common gaps in accessible video
Many publishers meet the letter of a requirement while missing its purpose. The most common gaps are unedited automatic captions, missing description of on-screen text, and flashing content that can trigger seizures in people with photosensitive epilepsy.
Another gap is important information shown only visually, such as a phone number in a graphic or a chart that the narrator never reads aloud. Writing scripts that describe key visuals in the main audio reduces the need for separate description and benefits all viewers.
The idea in brief
Accessible video is video that everyone can follow, including people who cannot hear, cannot see well or cannot use a mouse. It adds text for the spoken words, a narrated description of what happens on screen and controls that work with a keyboard or screen reader.
Think of it like a building with ramps, elevators and clear signs. The building is the same, but more people can get in. Many viewers without a disability also use these features, for example when watching captions on a quiet train.
A practical scenario
A city administration publishes a ten-minute video explaining a new waste collection system. Before release, the communications team commissions edited closed captions in WebVTT, an audio-described version in which a narrator reads the map and the collection calendar shown on screen, and a downloadable transcript.
The video player is tested with keyboard navigation and a screen reader, and one flashing transition is removed because it exceeded three flashes per second. The social media cut uses open captions because most viewers scroll with sound off.
After launch, residents who are deaf, blind or still learning the language can follow the same instructions as everyone else, and the administration meets its WCAG level AA obligation for public sector content.