Character recognition systems and hispanic printings of the Modern Age. The creation of two HTR models for the automatic transcription of documents in gothic and round scripts (XV-XVII Centuries)

Authors

  • Stefano Bazzaco Università di Verona https://orcid.org/0000-0002-3496-3687
  • Ana Milagros Jiménez Ruiz Universidad de Zaragoza
  • Ángela Torralba Ruberte Universidad de Zaragoza
  • Mónica Martín Molares Universidade da Coruña

DOI:

https://doi.org/10.13136/2284-2667/1190

Keywords:

Digital Humanities, Transkribus (READ Coop), HTR (Handwritten Text Recognition), early printed documents, Spanish Golden Age

Abstract

The work presents the recent achievements in the field of text recognition carried out in 2021 thanks to the collaboration between the following projects: Progetto Mambrino (Univ. of Verona), BIDISO (Univ. of A Coruña) and COMEDIC (Univ. of Zaragoza). Specifically, the first part of the article describes the state of the art of automatic transcription systems in relation to the recognition of printed texts of the Modern Age, the first experiences carried out with the Transkribus platform (READ Coop) and the preliminary results obtained. In the second part, we present two HTR models that allow the automatic transcription of early printed texts in gothic and round scripts of the Modern Age (15th-17th centuries). In two final appendices, the documents used for the creation of both models are described according to current typobibliographical standards.

Author Biography

  • Stefano Bazzaco, Università di Verona
    Dipartimento di Lingue e Letterature Straniere

Downloads

Published

2022-06-29