Archives and libraries hold large quantities of handwritten documents that are significant for research and cultural memory but are often difficult to access. Automated text recognition (ATR) can help address this issue. Algorithms and AI models convert handwriting and old printed text into machine-readable text, thereby enabling the analysis of large document collections.
The Potentials of Advanced Text Technologies (PATT) project had two objectives: to establish a panel of experts and to develop a learning module.
In collaboration with the ETH Library, the Swiss Social Archives, the State Archives of the Canton of Zurich, the UZH Archives, and the Zurich Central Library, a cross-institutional panel of experts was established. Regular workshops facilitated the exchange of current research findings and the further development of best practices.
The second outcome was a learning module on the use of ATR, which is freely available on the Ad fontes Open Access e-learning platform. It is aimed at researchers and interested laypeople and explains how ATR tools can be used in accordance with the current state of the art.
The expert group will be continued on a rotating basis by the participating institutions, and the Ad fontes platform is to be gradually aligned more closely with ATR in the coming years.