Skip to main content

Add Automatic Captions in Video on Demand

Video On Demand will automatically add captions to all newly-uploaded videos using Automatic Speech Recognition (ASR). ASR captions may be edited to meet WCAG AA guidelines. 

This article applies to: Video on Demand


Accessibility Requirements

To meet Federal ADA accessibility requirements, video content on webpages must be accessible to individuals with disabilities. This is provided by creating captions for video materials, so that deaf and hard-of-hearing users can make use of the content.

Cornell's Video On Demand service will automatically add captions to all uploaded videos using Automatic Speech Recognition (ASR). 

As a backup to the automatically generated captioning process, users can also manually start the process of generating ASR captions. The instructions below can be used to recreate captions that have been accidentally deleted, or to add captions if your video is otherwise missing them. Once captions have been manually requested, ASR will usually generate them within 10-20 minutes, though the process may require more time for longer videos, or when ASR is handling a heavy load of caption requests.

Automated Speech Recognition (ASR) captions are rated at 70-80% accuracy. This varies depending on the audio quality of the recording, the clarity of the speaker(s), and the amount of technical terminology used. Given these variables, ASR captions will not, by themselves, meet the accessibility standards that Cornell seeks to achieve.

Video owners are responsible for editing the ASR captions to meet the WCAG AA accessibility requirement for caption accuracy. Automatically generated captions that have not been reviewed and edited will not meet WCAG standards.

Order Professional Human Captioning

Some users may want assistance creating or editing captions that meet accessibility requirements or more accurately match the video content. Professional human captioning services deliver captions with an accuracy of 99% and are available for a fee.

Re-generating Automatic Captions

Video On Demand will automatically add ASR captions to all newly-uploaded videos using Automatic Speech Recognition (ASR). 
If you manually request ASR captions after they have been automatically created, nothing additional will happen.

The instructions below are only necessary if you have accidentally deleted your captions, or are otherwise missing them.

Order Replacement Captioning

  1. Log in to Video On Demand.
  2. Click My Media.
  3. Click the title or thumbnail image of the video.


     
  4. From the video’s page, go to the Actions drop-down menu to the bottom right of the video.
  5. Select Caption & Enrich from the drop-down list.


     
  6. Select the Source Media Language for your video, then click Submit.
Remember, when you have generated automatic ASR captions to your video, you must then edit your ASR captions to meet accessibility requirements.

Clean Up Automatically Generated Tags

After the request has been submitted, ASR captions will automatically be created for the video. Refresh the page with the video (after captions have completed) to review the captions.

The machine ASR system will also auto-populate the media listing with AI-generated tags based on the caption content. If you wish to remove incorrect tags:

  1. Click Actions to open the menu.
  2. Click Edit.
  3. In the Tags field, click the small x inside each incorrect tag to delete it.

  4. Click Save when you are finished removing incorrect tags.

Was this page helpful?

Your feedback helps improve the site.

Comments?