JavaScript is Disabled
Your browser's JavaScript functionality is disabled. It has to be enabled to use this function of ConfTool.
Here you can find information on how to enable JavaScript
If you have any problems, please contact the organizers at programme@ica.org.

Conference Agenda

Overview and details of the sessions of this conference. Please select a date or location to show only sessions at that day or location. Please select a single session for detailed view (with abstracts and downloads if available).

Only Sessions at Date / Time

Session Overview

Session

AI Godot: pipeline, challenges and expectations of automated Gothic document transcription

Time:

Thursday, 30/Oct/2025:

9:00am - 10:30am

Location: Room 123

Barcelona Supercomputer Center

Session Abstract

Autors: Coll Ardanuy, Mariona; Cuadrada Majo, Maria del Coral; Carrascosa, Adrián; Crosas, Mercè; Málaga, Alexandra ; Sarobe, R.; Berganzo-Besga, I. (ponent); Orengo, H.A. (ponent)

Until recently, large-scale text-based quantitative analyses were considered infeasible in medieval studies. This talk will show how recent advances in artificial intelligence have reached a satisfactory-enough level of readiness for new AI-based forms of data-driven historical research to be applied on large collections of digitized medieval manuscripts.
AI Godot is an interdisciplinary effort (archives, medieval history, computer vision, NLP) to unlock three centuries of Catalan notarial protocols (AHPB, 1297–1500; ~3,500 volumes, ~750,000 images) for large-scale, data-driven medieval studies. Our pipeline integrates document layout analysis, line segmentation, and handwriting text recognition tuned to Gothic scripts, followed by language identification and normalization (Latin/Catalan variants), and LLM-assisted post-correction, entity resolution, and information extraction.
This talk will aim to: a. explore the current challenges in automated medieval document transcription; b. provide details on the pipeline implemented to address these; c. describe expected results; and d. analyse its possibilities for large-scale deployment and adaptation to different corpora.
The talk aims to disseminate knowledge on the capacities and potential of the integration of NLP and AI for archive research. As such, the talk will be delivered in an open Q&A format and will be open to interruptions for questions, clarifications, and discussion.