In this session, we will demonstrate the Marriott Library's new AI tooling infrastructure that utilizes open source LLMs and ASRs to improve web accessibility and discoverability in the Digital Library at the University of Utah.
Universities, Schools, and Libraries are under growing pressure to expand access and improve discovery while meeting new accessibility expectations for our digital collections. At the University of Utah’s J. Willard Marriott Library, Digital Infrastructure Development team is building a flexible, modular workflow to help bring large AV collections into alignment with DOJ accessibility requirements at scale. The platform orchestrates open-source speech-to-text and language models to generate time-aligned transcripts and captions, structured segmentation, word clouds, entity recognition, and descriptive metadata—improving both compliance and content discoverability. We will share our implementation approach, early results, and lessons learned, including human review checkpoints, staff support and buy-in, provenance and auditability, and how we are designing the workflow to be adaptable as models and standards evolve. The session will focus on practical strategies other institutions can reuse to accelerate accessible AV delivery without locking into a single vendor or toolchain, and our future development plans for supporting other formats, including images, PDFs, and other formats.
