One Step Is Not Enough: A Multi-Step Procedure for Building the Training Set of a Query by String Keyword Spotting System to Assist the Transcription of Historical Document

Abstract

Digital libraries offer access to a large number of handwritten historical documents. These documents are available as raw images and therefore their content is not searchable. A fully manual transcription is time-consuming and expensive while a fully automatic transcription is cheaper but not comparable in terms of accuracy. The performance of automatic transcription systems is strictly related to the composition of the training set. We propose a multi-step procedure that exploits a Keyword Spotting system and human validation for building up a training set in a time shorter than the one required by a fully manual procedure. The multi-step procedure was tested on a data set made up of 50 pages extracted from the Bentham collection. The palaeographer that transcribed the data set with the multi-step procedure instead of the fully manual procedure had a time gain of 52.54%. Moreover, a small size training set that allowed the keyword spotting system to show a precision value greater than the recall value was built with the multi-step procedure in a time equal to 35.25% of the time required for annotating the whole data set.

End-to-End Transcript Alignment of 17th Century Manuscripts: The Case of Moccia Code

“…However, the digital transcript is not always linked to the manuscript image, making it difficult to locate the parts of the image that correspond to a particular part of the transcript. Thus, tools for transcript alignment, i.e., the automatic linking of the digital transcription with the parts of the digital image of the manuscript where it appears, would be of great help to scholars and historians who have to work with different versions of documents and/or writing styles, as well as for engineering a semi-automatic procedure for building the training set with minimal annotation, as suggested, for instance, in [ 8 ].…”

Experimenting with Training a Neural Network in Transkribus to Recognise Text in a Multilingual and Multi-Authored Manuscript Collection

“…Due to the rapid digitisation of archives, many projects have investigated the best practices to develop HTR models capable of automatically transcribing large collections of historical manuscripts (such as [17][18][19][20][21][22]). Most of these models prove to be very per-formative, having been trained on large corpora of coherent documents.…”