Unlocking the Digitized Archive of Early Modern Print

The Automatic Transcription of Early Modern Printed Books

Bibliographic Data

ID7167615
AuthorsSerena Strecker (Yale University), Kimberly Lifton (Yale University)
Year2025
Volume56
Issue2
Pages395-419
Publication date2025-06-01
Open AccessNo
TypeARTICLE
VenueSixteenth Century Journal (JOURNAL)
Journal identifiersISSN: 0361-0160 • E-ISSN: 2326-0726
PublisherUniversity of Chicago Press (PUBLISHER • US)
DOI10.1086/735052
OpenAlexW4410956017
LanguageEN
References cited7

Major digitization projects have made sixteenth- and seventeenth-century printed books more accessible than ever, yet most of these digitized sources have not been transcribed due to time, labor, and budget constraints. While earlier transcription projects relied on outsourced labor, equity and practicality demand a different workflow for obtaining transcriptions of digitized early modern sources. The new technology of Handwritten Text Recognition (HTR) promises to drastically reduce the human labor involved in transcription, yet how HTR should be integrated into the study of early modern sources has not been discussed broadly by the community of early modernists. This article argues that empowering individual researchers to build their own HTR models based on their specific corpus not only produces highly accurate transcriptions but also avoids unethical labor practices. A case study of four sixteenth-century German exempla collections shows even publicly available HTR models can produce highly accurate automated transcriptions of early modern printed books. Scholars who rely on early modern sources must decide how labor and technology should be combined to produce the transcriptions on which the next generation of early modern studies will depend

Information retrieval · Linguistics · World Wide Web · Computer Science · Digital Humanities and Scholarship · Handwritten Text Recognition Techniques · Image Processing and 3D Reconstruction · Philosophy

Tools

Open DOI