The Washington State Audio Description Project (WSADP), coordinated in Summer 2021 by The DO-IT Center at The University of Washington, was a project to support higher education institutions across Washington state in building their capacity to add audio description to public videos.
About Audio Description
Audio description is a narrative audio track that accompanies a video, describing important visual content for people who are unable to see the video. Individuals who are blind can understand much of a video's content by listening to its audio. However, if a video includes content that is only presented visually (e.g., on-screen text or key actions that are not obvious from the audio) this visual information must be described. This benefits people who are unable to see the video due to blindness or low vision. It also benefits users who are visually distracted, watching a video in the background while multitasking.
Project Participants
- University of Washington
- Washington State University
- Western Washington University
- Seattle Pacific University
- Olympic College
- Bellingham Technical College
- Whatcom Community College
- Shoreline Community College
Participants received:
- Guidance on how to prioritize videos for audio description, how to choose the best method of delivery, how to choose a vendor, and how to deploy the final product.
- Free audio description (we covered the cost of all videos described through the project).
- Technical support from our staff at any point during the process.
Participants agreed to:
- Work collaboratively with key stakeholders at their institution to identify the highest priority videos for description; to chose a method for delivering audio description for their chosen videos; and to explore how the effort to describe videos can be sustained after the project ends.
- Work collaboratively with project staff to add audio description to at least one public-facing video (the project covered the cost and project staff provided support as needed and served as a liaison with the audio description service provider).
Advance Research in AI for Audio Description
A team of students at the University of Washington was working to create a tool that uses artificial intelligence to automatically describe videos. All videos submitted through the current project, in addition to being described professionally, were additionally automatically described using the students' service. This helped them to advance the state of the science in this area, and potentially could have a significant impact on the availability and affordability of described video. Check out these examples of their work (all links open on YouTube):
For information about the students' findings, see the paper by Lucy Jiang and Daniel Zhu, VerbalEyes: A Large-Scale Inquiry into the State of Audio Description (in PDF). Since the conclusion of the current project, some members of the student team have launched VerbalEyes, a company that seeks to leverage technology to semi-automate the creation of audio descriptions in order to scale the production to match the growing demand.