Case Study 9.2: University of Texas, Austin, and Texas A&M International University—Captioning Workflows

In “Accessibility of Content in Institutional Repositories,” Colleen Lyon and Nerissa Lindsey recount strategies they pursued when working to improve the accessibility of media materials in their respective IRs.

Based at the University of Texas (UT) at Austin, Lyon began examining the accessibility of media materials in Texas ScholarWorks with assistance from a student at the university’s iSchool. After examining options for generating captions, Lyon and her team concluded that doing this work manually was too time-consuming to sustain going forward. They, therefore, requested $20,000 from UT Libraries to caption approximately 1,300 library-generated A/V materials. Upon receiving funding, they outsourced captioning work to on-campus media services.

The captioning project took approximately six to nine months to complete with additional time to upload captions and transcripts individually to the institutional repository. Subsequently, the ScholarWorks team shifted its policies to require captions and transcripts for all new deposits. Prior to receiving submissions, they actively point faculty, staff, and students to UT’s automated transcription tool and advise planning for transcription and captioning work from the beginning of media projects. The library’s event-planning group also builds in time and expense for captioning and transcription work when electing to record and deposit videos in the repository. ScholarWorks staff also began discussions with the UT Austin Graduate School about ensuring the accessibility of ETDs and supplementary files deposited by graduate students.

For examples of captioned videos in Texas ScholarWorks, see the Clifford Antone Lecture Series or Team Mission-Aransas – by the Reserve.

While previously based at Texas A&M International University (TAMIU), Nerissa Lindsey took on a similar captioning project for the TAMIU DSpace Repository. This work began with support from the Office of Information Technology (OIT), which began paying to have its media captioned by Rev.com at a rate of approximately $1 per minute. After completing captions, Rev sent SRT and TXT files to OIT, where a student used Submerge software to merge the SRT file with the MP4. The completed video files and transcripts were then sent to the library to be uploaded to the repository.

After this initial foray into captioning, the library funded captioning of archival videos in its International Bank of Commerce Keynote Speaker Series. The ultimate goal will be to caption all videos in the repository as funding is available.

Lyon’s and Lindsey’s projects underline questions to consider when initiating media accessibility projects:

• Which materials need remediation? How many items, and what type(s) of remediation work is needed?
• What tools are available to automatically or manually produce captions? Is it more cost-effective to do this work in-house or to outsource?
• What funds are available to support this work, and how much time can staff and students allocate to it?
• What units on campus might support this project with funds, expertise, or labor?
• What system constraints must be navigated to complete this project?
• How can policies be adjusted to ensure accessibility of all new deposits?