Multimedia accessibility
Last updated:
Learn how to ensure that all multimedia content, such as videos and audio files, is accessible for all users, including people with disabilities.
In this guide, we discuss the three required aspects of accessibility for pre-recorded multimedia: captions, transcripts, and audio descriptions. We’ll share information about how to create these materials yourself, use the university’s Kaltura MediaSpace tool, or send the multimedia to a UW–Madison approved vendor.
Captions, transcripts, and audio descriptions
Text alternatives for multimedia are essential for some individuals to access content, and can also help many others gain a deeper understanding of the material.
In this guide, we discuss captions, transcripts, and audio descriptions for pre-recorded material. To learn more about captioning and sign language interpretation for a live event, refer to our Live captioning and sign language interpreting guide.
Captions
Text version of the audio content within a video, synchronized with the sound on screen. Captions should be accurate, use correct punctuation, state in brackets who is speaking at the start of a new speaker, and indicate relevant ambient sound, music, or emotional tone.
Transcripts
Text version of media content, most commonly used for audio-only multimedia, such as a podcast. A transcript should capture all the spoken audio, plus background noise that wouldn’t otherwise be accessible without hearing the audio, like laughter and applause.
Audio descriptions
Audio track that describes and gives context for essential visual information in a video. By capturing what is happening on screen, audio descriptions make videos and other media accessible to people with low vision or who are blind.
Kaltura services
UW–Madison has contracted with Kaltura to automatically apply human-generated captions and machine-generated audio descriptions to videos uploaded to the Kaltura MediaSpace.
UW–Madison Kaltura MediaSpace
- The service automatically applies human-generated captions and machine-generated audio descriptions within four business days of being made “public” or “unlisted.” Learn more about How Human Captioning and Machine Extended Audio Descriptions Work.
- If you would like captions applied to videos that were already in Kaltura MediaSpace, including videos embedded in Canvas, prior to August 24, 2026, follow the steps for Ordering Human Captions for Media in Kaltura MediaSpace.
- Non-English captioning is available from other vendors at a separate cost, billable to the department ordering this specialty service. Find these rates in the Media captioning, transcription, and audio description vendor pricing KB.
Other vendors
For videos stored outside of Kaltura, UW–Madison has discounted rates with approved vendors Verbit and 3Play Media.
Verbit
- Verbit quick guide
- Set up service with Verbit
- Service includes audio descriptions and both English and non-English captioning and transcription
3Play Media
- 3Play quick guide
- Set up service with 3Play
- Service includes audio descriptions and both English and non-English captioning and transcription
For more information, visit Media captioning, transcription, and audio description vendor pricing.
Do-it-yourself methods
DIY captioning
Auto-generated caption tools are increasing in quality and accuracy. Some video hosting services, such as YouTube, support DIY captioning or the inclusion of separate caption files. Some web conferencing platforms allow you to download a transcript or caption file from a recorded event.
If you use auto-generated captions or transcripts, you must edit the file for accuracy before using it.
Check out Google’s captioning guide for YouTube or search online for guidance on how to create captions in the specific platform you’re using.
Follow these steps to add captions to your videos:
- Record a video. Try to capture high quality audio to increase the accuracy of auto-captioning tools.
- Upload the video to your preferred platform that allows generating and editing captions. Kaltura and YouTube are popular and user-friendly.
- Use the auto-captioning function or add a caption file that you already have.
- Review and edit the results for accuracy.
- Share the appropriately captioned video.
Captions and transcripts should include the following:
- Punctuation
- Speaker names in brackets when they begin talking
- Relevant ambient sounds in brackets
- Vocal or emotional tone in brackets, if needed
In addition to the requirements above, transcripts must include:
- Any on-screen text or visual information that is not read aloud in brackets
- Timestamps, when useful
W3C provides useful guides for captions, transcripts, and transcribing audio to text.
Webex
More information about Webex accessibility
Zoom
More information about Zoom accessibility
Microsoft Teams
More information about Teams accessibility
DIY audio description
Audio descriptions are verbal explanations of what’s happening on screen and are useful for people who are blind or have low vision. Audio descriptions can be strategically added during natural pauses in dialogue. Here are tips to help you create your own audio description.
Describe what is happening on screen in the video. Focus on:
- Characters and people
- Action or movement
- Charts, diagrams, and other images
- Scene and scene changes
When recording audio descriptions:
- For new content, create integrated (embedded) audio descriptions, which is when narrators or speakers in the video describe visuals as they appear on screen. If you’re writing a script, write these visual descriptions into the narration script.
- For pre-recorded content, try to describe visual content during space in audio, ideally as or just before the content is presented. If there is not sufficient space in the audio, space may need to be created by inserting or pausing on a static image.
For information on how to add audio descriptions, read about the audio description of visual information or see the Described and captioned media program’s Description Key.
Embedded videos
In addition to captions and audio descriptions, it’s important to remember the following characteristics of accessible embedded videos:
- Playback controls must be accessible via keyboard and screen reader
- Videos should not automatically play upon loading the page
Video hosting platforms might have options when creating the code to embed a video. When available, select options to include video controls and to turn off auto-play.
To ensure accessibility for keyboard and screen reader users, make sure that media players support the following interactions:
Keyboard accessibility
- Controls can be accessed using the Tab and Arrow keys and there is visual indication of which control has focus
- Use Enter or Space Bar to Play/Pause
- Use Arrow keys to control volume, timeline buffer, and Forward/Rewind bars
Screen reader accessibility
The name, state, role, and value of controls is announced to the end user when applicable. For example:
- “Play button”: The name (Play) and role (button) give the user enough information to play and pause the video.
- “Captions toggle button, selected”: The name (Play), role (toggle button), and state (selected) tell the user how to toggle captions and whether or not they are selected.
- “Video timeline slider, 0:30 of 2:30”: The name (Video timeline), role (slider), and value (0:30 of 2:30) tell the user how to interact with the timeline and where they are in the video.
Get help from the Center for User Experience (CUE)
We partner with you to create accessible, usable, and inclusive digital spaces through free evaluations and consultations.