AMIA 2024

Since the 2022 debut of Whisper, OpenAI’s automatic speech recognition software, an ecosystem of complementary tools and modifications has evolved, now offering organizations stewarding AV materials unprecedented opportunities to leverage this tool to increase the accessibility and discoverability of their digital collections. In this session, panelists from Emory, GBH, and Indiana will discuss their experiences and research implementing Whisper into their media management and preservation workflows, discussing Whisper’s performance across varied AV collections as well as the human impacts of working with Whisper. These insights from three organizations with varied scopes of collections and distinct but connected avenues of research may be helpful to a variety of organizations engaged in similar work or interested in starting up such a project. Attendees will gain insight into the strengths and weaknesses of leveraging Whisper, the challenges and opportunities presented by the technology, and practical guidance on implementing or expanding AV accessibility-related projects.