Advancing Medieval Manuscript Transcription: Insights from the Vienna HTR Winter School

2 February 2025

CDH/MARBAS Postdoctoral Research Associate Christine Roughan shared her experience at the Winter School on Handwritten Text Recognition (HTR) of Medieval Documents in Vienna, Austria.

A group of people stand on a staircase

Participants in the HTR Winter School recently participated in a three-day workshop in Vienna, Austria.

In December, CDH/MARBAS Postdoctoral Research Associate Christine Roughan traveled to Vienna for the three-day workshop concluding the Winter School on Handwritten Text Recognition (HTR) of Medieval Documents, organized by MARBAS, the Institute for Medieval Research at the Austrian Academy of Sciences, and the Faculty of Arts at Comenius University.

The workshop brought together over 100 participants from twenty-two countries eager to build upon the knowledge they gained in the three Zoom sessions, which provided an introduction to HTR technology that enables computers to produce transcripts of manuscripts. Participants joined one of six groups studying languages from Carolingian Latin to Medieval Czech to Byzantine Greek.

In addition to co-facilitating the Syriac language group with Ephrem Aboud Ishac (Austrian Academy of Sciences), Christine presented to the whole group on “HTR Platforms: eScriptorium & Transkribus.” As Christine explained, the presentation introduced students to a wider range of HTR tools and showed them “how they could leverage existing datasets,” including those produced during the workshop, “in a range of tools.”

We caught up with Christine to learn more about her experience with Winter School.

How did you become involved in Winter School?

In the summer of 2024 I was invited to the Institut für Mittelalterforschung at the Austrian Academy of Sciences to give a talk, “Computational Approaches to Textual Multiplicity in Curricular Manuscripts.” I became acquainted with many of the organizers then, and as my talk involved handwritten text recognition we came to discuss the Vienna HTR Winter School. This past fall was therefore my first year being involved.

Why is the Vienna HTR Winter School important?

There is an absolute wealth of historical textual material stored in libraries and archives around the globe, and for many years now we have seen institutions pursuing digitization initiatives that have made the contents of their collections more widely available. These initiatives produce digital images—the next step, getting a searchable or otherwise machine-actionable text out of those images, requires either manual transcription or a technology like handwritten text recognition (HTR).

Many of us are already used to computers being able to transcribe images of printed text (OCR)—today the technology has made great strides to the point where we are able to use HTR workflows to get a fairly accurate text out of, say, a medieval manuscript! So at the Center for Digital Humanities, we have been seeing a surge of interest from Princeton community members who are interested in learning how to apply this technology to their own projects with manuscripts, archival documents, and other handwritten materials. Opportunities like the Vienna HTR Winter School serve as a fantastic space for students and scholars to gain intensive training and experience with these tools and workflows.

How did your involvement advance your own research and teaching?

My own research with medieval Arabic and Greek manuscripts benefits from HTR and so I came to Vienna with no lack of practice. My time with the Winter School has furthered my teaching in the subject and my experience with newcomers to HTR, both experiences which I am bringing back with me as I continue my involvement in co-leading the Princeton Open HTR Initiative with Princeton faculty members Helmut Reimitz (History) and Marina Rustow (Near Eastern Studies and History).

For more information on Winter School, visit https://www.oeaw.ac.at/imafo/das-institut/detail/handwritten-text-recognition-of-medieval-documents.