An Evaluation of GPT-4V for Transcribing the Urban Renewal Hand-Written Collection
Lee, Myeong; Hsu, Julia Hsin-Ping
George Mason University, United States of America
HTML XMLBetween 1960 and 1980, urban renewal transformed many cities, creating vast handwritten records. These documents posed a significant challenge for researchers due to their volume and handwritten nature. The launch of GPT-4V in November 2023 offered a breakthrough, enabling large-scale, efficient transcription and analysis of these historical urban renewal documents.
HTR to TEI: Strategies for Transformation, Up-conversion, and Data Enrichment of Transcripts
Cummings, James (1); Jakacki, Diane K. (2)
1: Newcastle University, United Kingdom; 2: Bucknell University, USA
This short paper originates from both the Evolving Hands project and data processing experience of the authors. The topic is not the project itself, or the process of Handwritten Text Recognition (HTR), but the transformation, up-conversion, and enrichment of the output data. It targets those considering HTR for TEI output.
A quantitative and qualitative evaluation of large scale handwriting recognition models for Norwegian
Tarride, Solène (1); Beyer, Yngvil (2); Roald, Marie (2); Enstad, Tita (2); Boillet, Mélodie (1); Kermorvant, Christopher (1)
1: TEKLIA, France; 2: National Library of Norway
In the ongoing Hugin Munin project, our aim is to achieve high-quality Handwritten Text Recognition (HTR) for the majority of documents in the National Library of Norway. For this purpose, we introduce the NorHand dataset, comprising annotated Norwegian documents. We train and evaluate three state-of-the-art models for Handwritten Text Recognition.
Responsible transcription and text markup practices to enhance accessibility for people who use screen readers
Van Hyning, Victoria (1); Jones, Mason (1); Jordan, Bern (1); Mahmood, Zuhair (2)
1: University of Maryland, iSchool, United States of America; 2: US Government Accountability Office
HTML XMLThrough a case study of crowdsourced transcriptions, we argue that cultural heritage and DH practitioners and scholars should prioritize enhancing textual data accessibility. We center the experiences of disabled users whose access to transcriptions in cultural heritage collections is woefully limited by poor or non-existent markup, and confusing search infrastructure.
Mapping Latin American Women’s Intellectual Networks
Duarte, Diana Milena
Emory University, United States of America
"Mapping Latin American Women’s Intellectual Networks" uses data from 23 newspapers to visualize 19th-century networks of women writers. It categorizes biographic, bibliographic, temporal, and spatial data, resulting in a collection of 1,000 texts by 50 women from 1830 to 1900. This initiative aims to centralize and make accessible this information.