Jan Thiele from the Consejo Superior de Investigaciones Científicas (CSIC), Madrid, will present on Advances in Developing Arabic HTR.

The textual production of cultures utilizing non-Latin scripts remains vastly understudied compared to the Latin tradition. It is fair to say that more than half of the surviving texts remain in manuscript form, severely limiting accessibility and creating substantial gaps in our understanding of these cultures. If traditional methods of exploration, transcription, and editing of manuscript texts continued—namely, manual approaches—we would likely require another century to obtain a reasonably complete picture of textual production in non-Latin scripts. However, recent advancements in technologies aimed at automating the recognition and transcription of handwritten text (HTR) offer promising prospects for a more efficient and less biased exploration of past cultures. Focusing on recent advances of Arabic script HTR, this presentation offers a critical evaluation of the latest technologies. It addresses whether they have the potential to effectively overcome the longstanding bias toward the study of cultures using non-Latin script. It specifically examines the sustainability of these technologies, assessing criteria such as their resource intensity (in terms of both technical requirements and computational know-how), as well as their compliance with principles of open science (including software licensing, and the production and distribution of FAIR data, etc.). 

The Research Colloquium led by Prof. Dr. Beatrice Gruendler regularly features international guest speakers in the field of Arabic Studies and related disciplines. Colloquium sessions are being held in a hybrid fashion, combining personal encounter with the advantages of remote access. The shared language is English.

Students of Arabic Studies or related disciplines, visiting scholars, and colleagues are cordially invited to participate, also for single sessions.