Captions, transcripts and audio descriptions make multimedia accessible to those with vision and hearing impairments, and learning disabilities. These also assist people viewing videos in noisy environments, those who have difficulty concentrating, textual learners, individuals not fluent in English, and others.
- Transcript: A written or printed version of material originally presented in another medium. Transcripts allow users to search for specific information or keywords contained in the media.
- Closed Captions/Subtitles: A text-based description of the video’s dialogue including speaker identification and other relevant audio information (i.e., music or sound effects). The text typically appears on the bottom of the screen but may be placed in different locations. Captions are synchronized with the media.
- Automatic Speech Recognition (ASR) or automatic captions are machine-generated and require human editing to ensure accessibility compliance.
- The most common use for subtitles is for translating dialogue into another language.
- Communication Access Realtime Translation (CART): CART is the captioning of live events. See WebAIM’s page on Real-time Captioning.
- Audio Descriptions (AD): An additional audio track or narration that describes and gives context for essential visual information in media and live productions.
- The university is required to provide equally effective communication for people with disabilities and seeks to meet and exceed the requirements of the Americans with Disabilities Act, Sections 504 and 508 of the Rehabilitation Act and state law. See Accessibility Laws, Policies and Standards. To provide equal access, staff should provide accurate captions for all video.
- For requests for student accommodations, contact the Disability Center.
- For audio-only content, such as a podcast, provide an accurate text transcript.
- The text transcript must be an accessible document, preferably HTML.
- If you provide a text transcript in other formats, such as a .TXT file, the transcript/link must be under or close to the audio player.
- Automatic captioning must be reviewed for accuracy by a human.
- For live video, provide live captions or a prepared transcript, statement or script that contains equivalent information to what is spoken.
- Provide a way for viewers to turn off sound that automatically plays for more than 3 seconds.
How to Add Captions and Transcripts
Creating accurate closed captions and transcripts to meet accessibility compliance requires specific knowledge and takes time. Hiring a professional video captioning service is cost-effective, provides a quick turnaround and meets compliance.
If you choose to caption videos yourself, see the tutorials and free ASR captioning tools listed below. You must edit these machine-generated captions for accuracy to meet compliance.
- Industry Standards
- Making Audio and Video Media Accessible
- Captions, Transcripts and Audio Descriptions
- Video Captions & Transcripts
- Create Accessible Video and Audio
- YouTube Captioning
- Caption and Description Editing Tool
- Amara Captioning
- Working from a script will make it easier for you to create a transcript or captions
- Include descriptions of any images if the speaker does not fully explain what is presented
- Creating Accessible Videos
- Captioning Style Guide
Panopto: recording lectures
- Captions can be automatically generated
- You can also upload a caption file
- Panopto Accessibility Features
- Panopto FAQ
- Panopto support
Kaltura: video platform
If you are hosting videos in the Kaltura Media Gallery (e.g., in a Canvas course site), you will need to add the captions.
VoiceThread is an application integrated with Canvas that allows users to create presentations, post comments, and share content.
Zoom: video conferencing
If you record a Zoom session, a transcript will be generated. The transcript must be edited for accuracy. You can then use the transcript to add captions when the recording is shared in the Canvas Media Gallery.
The following vendors are contracted to provide human transcription, captioning and CART services. To get started, contact any vendor listed below. Contact Procurement for questions.
The Data Classification Level (DCL) of the data being captioned, transcribed or made audible in accordance with this contract must be limited to DCL1 (public data) and DCL2 (sensitive data) data. The use of these services for DCL3 (restricted) or DCL4 (highly restricted) data is not approved by the university. Refer to the university’s Data Classification System for more information.
While units are not required to select a contracted vendor, they are highly encouraged to do so. By choosing one of the vendors, you ensure projects will adhere to all university standards and requirements — including the IT Compliance standards — and eliminate the need for a formal RFP, significantly expediting your timeline.
If you do not use one of these vendors, the purchasing guidelines apply. See the Procurement Guidelines.