Increasing Accessibility of Audiovisual Content Using Whisper : Poster - iPRES 2025 - Wellington, New Zealand / Te Whanganui-a-Tara, Aotearoa
Abstract
Captions are essential to the searchability, discoverability, and accessibility of audiovisual (AV) materials. However, creating captions is often challenging due to the resources required. With data collected from a grant-funded project, this poster describes the impact of Whisper [1], an open-source automated speech recognition (ASR) software, on captioning work at Emory Libraries, observing that Whisper increased capacity for captioning without entailing a reduction in staff time, due to increasing demand and the ongoing need for human review in providing accurate, high-quality captions. The project methodology and findings will be presented along with implications for digital preservation and leveraging ASR-generated captions in metadata enhancement.Details
- Creators
- Nina Rao; Simon O'Riordan
- Institutions
- Date
- 2025-09-01 00:00:00 +0000
- Keywords
- digital humanities
- Publication Type
- poster
- License
- http://creativecommons.org/licenses/by-sa/4.0/
- Download
- 149717 bytes