Next Article in Journal
Characterization and Analysis of the Mortars of the Church of San Francisco of Quito (Ecuador)
Next Article in Special Issue
Potential Impact of Using ChatGPT-3.5 in the Theoretical and Practical Multi-Level Approach to Open-Source Remote Sensing Archaeology, Preliminary Considerations
Previous Article in Journal
Yellow Dyes of Historical Importance: A Handful of Weld Yellows from the 18th-Century Recipe Books of French Master Dyers Antoine Janot and Paul Gout
Previous Article in Special Issue
A Quantitative Social Network Analysis of the Character Relationships in the Mahabharata
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Heritage
Volume 6
Issue 12
10.3390/heritage6120392
heritage-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Experimenting with Training a Neural Network in Transkribus to Recognise Text in a Multilingual and Multi-Authored Manuscript Collection

by
Carlotta Capurro
1,*,
Vera Provatorova
2 and
Evangelos Kanoulas
2
1
Department of History and Art History, Utrecht University, Drift 15, 3512 BR Utrecht, The Netherlands
2
Informatics Institute, University of Amsterdam, Science Park 900, 1012 WX Amsterdam, The Netherlands
*
Author to whom correspondence should be addressed.
Heritage 2023, 6(12), 7482-7494; https://doi.org/10.3390/heritage6120392
Submission received: 4 September 2023 / Revised: 10 November 2023 / Accepted: 20 November 2023 / Published: 29 November 2023
(This article belongs to the Special Issue XR and Artificial Intelligence for Heritage)
Browse Figures
Versions Notes

Abstract

This work aims at developing an optimal strategy to automatically transcribe a large quantity of uncategorised, digitised archival documents when resources include handwritten text by multiple authors and in several languages. We present a comparative study to establish the efficiency of a single multilingual handwritten text recognition (HTR) model trained on multiple handwriting styles instead of using a separate model for every language. When successful, this approach allows us to automate the transcription of the archive, reducing manual annotation efforts and facilitating information retrieval. To train the model, we used the material from the personal archive of the Dutch glass artist Sybren Valkema (1916–1996), processing it with Transkribus.
Keywords: handwritten text recognition (HTR); neural networks; digital archives; digital cultural heritage; Transkribus handwritten text recognition (HTR); neural networks; digital archives; digital cultural heritage; Transkribus

Share and Cite

MDPI and ACS Style

Capurro, C.; Provatorova, V.; Kanoulas, E. Experimenting with Training a Neural Network in Transkribus to Recognise Text in a Multilingual and Multi-Authored Manuscript Collection. Heritage 2023, 6, 7482-7494. https://doi.org/10.3390/heritage6120392

AMA Style

Capurro C, Provatorova V, Kanoulas E. Experimenting with Training a Neural Network in Transkribus to Recognise Text in a Multilingual and Multi-Authored Manuscript Collection. Heritage. 2023; 6(12):7482-7494. https://doi.org/10.3390/heritage6120392

Chicago/Turabian Style

Capurro, Carlotta, Vera Provatorova, and Evangelos Kanoulas. 2023. "Experimenting with Training a Neural Network in Transkribus to Recognise Text in a Multilingual and Multi-Authored Manuscript Collection" Heritage 6, no. 12: 7482-7494. https://doi.org/10.3390/heritage6120392

APA Style

Capurro, C., Provatorova, V., & Kanoulas, E. (2023). Experimenting with Training a Neural Network in Transkribus to Recognise Text in a Multilingual and Multi-Authored Manuscript Collection. Heritage, 6(12), 7482-7494. https://doi.org/10.3390/heritage6120392

Article Metrics

Back to TopTop