General considerations

Audio needs to be clear; loud background sounds make it difficult for those who have hearing impairments to understand what is being communicated.

Don’t use flashing visual effects in videos as it can cause seizures in some users. Also, avoid excessive motion or high frame rates in video content as it can disorient users

Captions and Transcripts

Captions are synchronized with multimedia and are essential for those with hearing impairments. Captions are needed for pre-recorded and live media (including videos). All live audio-only content that has narration or dialogue must be accompanied by text-based synchronized captions.

Transcripts are the full text of spoken words and important visual information in the multimedia file. Transcripts are usually read as an alternative way of watching or listening to the content. Pre-recorded multimedia, audio-only or video-only content must include a text transcript.

There are similarities as to what captions and transcripts should include. Both need to:

• Be verbatim for unscripted or live content
• Use [brackets] or (parentheses) to convey important background sounds
• Capture speech that is spoken off-screen
• Identify the person speaking
• Use punctuation to convey emphasis wherever possible rather than writing more text to explain the emphasis
• Cannot reveal information ahead of planned audience communication; also cannot withhold information that has been communicated
• Identify music by title and artist; the music lyrics should be included, unless contextually inappropriate
• Use neutral language for speech that is inaudible or difficult to perceive clearly
• Refrain from editing out strong language, unless the actual version bleeps it
• Indicate when speech is whispered or mouthed

There are also several elements that apply only to Captions, or only to Transcripts.

• For instance, while captions need to be verbatim, there is a little leeway if simplified captioning would be more relevant for the target audience such as those with cognitive disabilities.
• Captions need to use conventional spelling except where phonetic spelling is essential to the content.
• Captions should also describe sounds, not the actions that cause the sounds.
• Important visual events need to be described on a transcript.

So how should captions be presented? I’ve included a list of what to keep in mind when showing them.

• No more than three lines on a screen at any one time
• Use logical points between phrases to insert line breaks
• Type using mixed case
• Use Sans Serif font
• No more than 32 characters per line of captions
• Present on a screen for a minimum of 1 second, and each word should get an additional 0.3 second to read. Additional time should be added if words are unfamiliar or uncommon. Also if a lot is going on in the frame more time for the caption should be given
• Shouldn’t present captions so that they overlap screen text, faces, or other visual elements
• Be precisely synchronized to the audio, unless it makes them unreadable
• Use white 22-point minimum font on a black background. 18-point font minimum is allowed but must have a colour contrast of 3:1.
• Don’t bold the text, or use colours to convey meaning within captions
• Word capitalization/Italicization can be used only if punctuation is not enough emphasis.
• Quotations and mixed-use capitalization should be used to designate titles of books or movies.
• Caption frames need to be removed from the screen for long silent intervals
• Gaps between captions need to be at least 1.5 seconds
• The visual appearance of captions should be customizable by users
• Captions need to be available in multiple file formats, at least one of which should be in a WebVTT format

Audio Descriptions

Pre-recorded multimedia (video and audio) and video-only content must have audio descriptions. It’s also a good idea to include them for live multimedia and video online content. Use extended audio descriptions for pre-recorded multimedia when pauses in the foreground audio are insufficient to allow audio descriptions to convey the sense of the video.

Sign language interpretation should be provided for all pre-recorded and live multimedia content.

Media Player Accessibility

Any media player that is used needs to allow users to access captions, transcripts, and audio descriptions at any time. They must allow for:

• The visual customization of captions
• Remembering preferences
• Providing full-screen functionality, and
• Being accessible to both keyboard-only and those who use screen readers

Background Sounds

During (pre-recorded or live) audio-only and multimedia content, background sounds need to be minimized (20dB lower than foreground sounds, except for occasional sounds lasting less than 2 seconds) or eliminated during narration or dialogue, or eliminated.

For background audio on web pages, there needs to be a mechanism to stop, mute, adjust volume, or pause audio that automatically plays for longer than 3 seconds on a page.

Flashing Content

Multimedia content must not contain visuals that flash more than 3 times per second, as these can cause seizures. Before uploading any content that you are unsure about, I strongly encourage the use of the Harding Flash and Pattern Analyser, to ensure content does not cause photosensitive epilepsy.

Background Videos

Full user playback control, along with captions, transcripts or audio descriptions must be available for any background video that conveys important content. Background videos must not have sounds, and movement must be minimal. Any overlaid text must have a minimum colour contrast of 4.5 to 1 for small text and 3 to 1 for large or bold text.

Auto-Play

Any content that automatically starts playing and lasts for 5 seconds or longer must allow a user to pause, stop, or hide it.