Informational and instructional videos used at the college level benefit from the use of human-corrected captions. Human-corrected captions make videos more accessible to students who are deaf or head of hearing, and universities like UNLV must ensure that more students have equal access to educational media and content by prioritizing multimedia accessibility. Overall, creating accessible captions is the right thing to do.

Defining Accessible Captions

According to the National Association of the Deaf:

Captions not only display words as the textual equivalent of spoken dialogue or narration, but they also include speaker identification, sound effects, and music description. It is important that the captions are:

  1. Synchronized and appear at approximately the same time as the audio is delivered
  2. Equivalent and equal in content to that of the audio, including speaker identification and sound effects 
  3. Accessible and readily available to those who need or want them

Captions must have sufficient size and contrast to ensure readability, and be timely, accurate, complete, and efficient. When displayed, captions must be in the same line of sight as any corresponding visual information, such as a video, speaker, field of play, activity, or exhibition.

Disclaimer About Auto-Generated Captions

Auto-generated captions, such as those produced when initially creating or uploading a video to Panopto or YouTube, are inadequate. In a settlement delivered in November of 2019, the court settled a lawsuit filed by the National Association for the Deaf against MIT and Harvard. This resulting consent decree informs the practices and standards by which universities like UNLV approach what is called human-vetted or human-corrected captions.

Characteristics of Conformant/Compliant Captions

Captions for content that is communicated, managed, or overseen by a subject matter expert (such as a course designer, instructor, or professor) should reflect the following:

  • Be 99% accurate. This includes spelling, grammar, punctuation, sentence structure, and lower/upper case letters.
  • Not exceed three lines on the screen. This means that there must only be one to three lines of text on screen at a time.
  • Coincide with or be synchronized with the audio track to maintain the relation between the visuals and the sounds.
  • Reflect slang words spelled verbatim.
  • Include non-speech sounds (such as background noises, sound effects, music, and music-track references) in square brackets.

Captions that are made or edited to reflect these qualities are called “human-corrected captions.” These characteristics contrast with the autogenerated AI captions from Panopto, WebEx, Zoom, YouTube, and other content distribution services.

Resources for Captions (Including Best Practices)

Determining When Captions Are Required

A caption decision tree to determine if you need to have human corrected captioning in your course

The conditions for which captions must be provided appear in the "Yes" portion of the above flowchart.

To be clear, an equitable class experience requires that the content in question is only presented when the content has been completely adjusted or corrected. The content should not be presented in stages of accessibility fixes or adjustments. 

Items That Do Require Captions

  • All new course developments
  • Subject matter expert self-generated content, such as Panopto, that has been recorded at an earlier time, with the intention or plan to be deployed once or multiple times in the future
  • Materials that will be reused in future courses
  • Captioning required due to a Disability Resource Center (DRC) request for a student in the course.

Items That Do Not Require Captions

According to the standards and procedures that support the UNLV ICT Policy, captions are not required when all four of these conditions exist or are met: 

  1. Subject matter expert self-generated content, such as Panopto 
  2. Content is for a currently running course 
  3. No DRC request 
  4. Content will not be reused in future courses

Captioned Videos from UNLV Libraries

The UNLV Libraries provide access to thousands of captioned streaming videos for UNLV use. For more information, please visit their Online Movies & Streaming Video page.

Getting Your Captions Fixed

UNLV currently has 3PlayMedia as its approved campus-wide contractor for captions and transcripts. If you wish to outsource caption generation, contact the Accessibility Resources Team Support. The Office of Accessibility Resources oversees UNLV accounts for 3PlayMedia. They can start the process for your college, department, program, or unit to control and maintain your captioning and transcription projects directly.

If you wish to employ an in-house support model using student workers, designated staff, or administrative faculty, OAR can supply consulting and resources upon request.