Transcribathon is a crowdsourcing platform that offers its users the unique opportunity to work with historical material and become an integral part of their digitisation process. This digital material — scanned images of historical letters, diaries and pictures — has been collected by various cultural heritage institutions and are showcased in our projects. By discovering, reading, and adding transcriptions, tags and georeferences to these documents, users play a vital role in increasing their access and open up their future potential for research and education.

The platform also combines Artificial Intelligence, such as Handwritten Text Recognition technology and contributions from members of the public acting as ‘Citizen Scientists’. The project also involves applying natural language processing and big data analysis technology to analyse transcriptions and their translations, providing support for semantic metadata enrichment, clustering and classification.