Snap. Flash. Smile.

Created by Dr. Joshua Miele and his colleagues at the Smith-Kettlewell Video Description Development Center in San Francisco, California, the Descriptive Video Exchange (DVX) provides an audio descriptor narrating the scene or moment in the program.

DVX is still in early development where large load balancing and optimization of geographic use doesn’t require multicloud use at this time. However, Miele noted it’s called cloud-based because it helps people understand that the project is all on the Internet and not a program that runs on one’s PC or smartphone.
What’s important to note is that the server itself does not store the original videos; rather, it holds only the descriptions, identification and timing information, and works in time with a video player that plays whatever video you want to watch.

At present, those who are visually impaired are able to access videos, such as movies or television shows, and listen to the dialog and music it contains. However, what remains missing is a certain visual context that is currently unavailable for most videos.

“We are looking into the future, where the vast majority of video will not be consumed via broadcast media. It will be consumed over the Internet,” he explained. “We are looking at a day coming very soon when electronic textbooks will contain embedded video.”

Though video description was invented in the 1970s as a way to give blind people access to information on what’s occurring digitally on the screen, it was mostly for entertainment.

However, video description isn’t only about the luxury of entertainment now, said Miele.
“Now there’s a great deal of education and job training and healthcare-related information distributed to video,” he explained. “If you don’t have access to the information on the screen, blind people won’t be educated or have the same access to job training as sight people do.”

Miele points to websites, including YouTube, where people can upload their own videos that have changed the landscape of accessibility for those with visual disabilities.

In order to create a described video, users would previously have to get permission from the copyright holder, he added.

“We’re interested in the idea of crowd-sourcing video description so that anybody anywhere can describe any online video for anybody anywhere else,” said Miele, adding they are currently looking into the legal and fair trade aspects of the project. “We aren’t redistributing or copying the video, we’re not doing anything to the video other than playing with it simultaneously with the described content.”

DVX is currently available upon request.

Along with creating DVX, Miele is working with his partners to create youdescribe.org, which will provide audio descriptors for popular YouTube videos.

The site itself has a little over 100 videos currently available to watch with audio descriptors, but it allows the public to record their own audio descriptors before sharing the result to the public.

Here′s a sample of a descriptive video.