Auto-Captioning for Kaltura

Abstract collage: Captioning
February 6, 2018

Videos uploaded to Kaltura in Moodle and Canvas now have auto-captions.

The new auto-captioning feature uses the Cielo24 Automatic Speech Recognition (ASR) service. This means that now when a viewer clicks the "cc" icon in the Kaltura player window while in Moodle or Canvas, they will be able to see closed captioning on any video uploaded after February 1, 2018.

This feature is in place for any instructor, staff, or student that uploads to Kaltura, and there are no settings for them to change to enable the captioning process.

Only 80% Accurate

While ASR captions are a step forward, it is important to know that they only offer around 80% accuracy. That means:

  • 1 in 5 words is incorrect
  • A 8-word sentence will be ≅17% accurate
  • A 10-word sentence will be ≅11% accurate

The WCAG 2.0 caption definition prescribes that the captions provide "information needed to understand the media content" and that "captions convey not only the content of spoken dialogue, but also equivalents for non-dialogue audio information".

Accurate Captions Help Ensure Inclusiveness

Accurate captioning is the only way to ensure that people who are Deaf and hard of hearing can understand audio content. In fact, most accessibility advocates would argue that automatic generated captions are actually detrimental to accessibility.

Using Auto-Captions as a Starting Point

However, auto-captions can be very useful as a starting point for providing captions. Learn how to Review and Edit Captions from Cielo24.

Need Help?

The Multimedia Hub is available to assist the UMD community in providing videos inclusive to all. The Hub is happy to work with you to make accessible videos.

Funding

As stated in UMD's Captioning of Video Policy, funding is available specifically for credit courses with an enrolled student who has a disability (documented with Disability Resources) which would be accommodated by captioning.

Further Information