Comparing human and machine performances in transcribing 18th century handwritten Venetian script [DH2018 Mexico City, 26-29 June 2018]

« Automatic transcription of handwritten texts has made important progress in the recent years (Sanchez et al., 2014; Sanchez et al., 2015, Sanchez et al., 2016). This increase in performance, essentially due to new architectures combining convolutional neural networks with recurrent neutral networks, opens new avenues for searching in large databases of archival and library records. This paper reports on our recent progress in making million digitized Venetian documents searchable, focusing on a first subset of 18th century fiscal documents from the Venetian State Archives (Condizione di Decima, Quaderni dei Trasporti, Catastici). (…) »