Technology from FIT can be used to convey personal testimony from the First World War

Diaries, memories and correspondence from the First World War will be conveyed to researchers as well as the public by the EGO-DOK, a new project of the Military History Institute Prague. After over a century, personal testimonies of both participants in and observers of the war will come to life in a digital world also thanks to technology in development at the Faculty of Information Technology.

It is the goal of the EGO-DOK project for the coming years to digitise historical documents from the First World War, especially documents of a personal nature, and concentrate them under a common search engine. The aim is to gather as many sources, which are currently scattered over many institutions, private collections and families, as possible, and preserve them for the future or make them available for researchers. The Military History Institute will now offer institutions and private individuals to process their documents into a digital form free of charge.

After scanning the documents, it will be a time for tools developed by FIT researchers within the PERO project. They use a unique technology for reading manuscripts utilising artificial intelligence and machine learning. "Our tools then completely process the document. They can locate lines, create an accurate transcription of texts written in both the Latin and Kurrent script and subsequently allow for full-text searching, including a function for highlighting the searched expression in the text," explains Martin Kišš from FIT who personally participated in the development of the technology.

The Military History Institute then hands the results of processing to the document's owner and publishes them in the Digital Reading Room of the Ministry of Defence of the Czech Republic. Even now the reading room offers the first fruits of yielded by the co-operation between the Military History Institute and FIT BUT - an example of a processed diary from the First World War period. 

Tools from the PERO project will also be used for example to refine searching of the contents of old newspapers, Czech cantastoria written in Fraktur and to transcribe a large amount of handwritten chronicles from the 20th century.