From Manuscript to Data: An Integrated Pipeline for Handwriting Recognition, Editing, and Indexing

The e-NDP ANR project aims to renew our understanding of Notre-Dame de Paris through the digital edition of the cathedral chapter’s registers. Twenty-six registers, compiled between 1326 and 1504, provide access to all decisions made by the canons regarding the administration of the cathedral, the management of their rights and property, and the society of the cloister. This article provides a concise overview of the processing pipeline developed to automate the reading and linguistic annotation of this extensive medieval corpus. With the aim of facilitating navigation through heterogeneous data collections, we have built an application designed as a central access point to all the resources generated. The e-NDP application for browsing and mining the corpus is presented here with an emphasis on the choices made to facilitate data reuse.