This AI system uses Transkribus technology to automatically transcribe historical Canadian documents, including newspaper indexes and government Order-in-Council Registers from 1925 to 1940. The system employs Intelligent Character Recognition (ICR) and Handwritten Text Recognition (HTR) models developed by READ-COOP to convert images of handwritten and printed text into searchable digital text. This pilot project, now retired, was developed by Library and Archives Canada in collaboration with the EU-funded READ project to improve public access to historical records and enable better digital preservation.
The system processes historical records that may contain personal information about individuals mentioned in the documents. The transcription results are made available to both Library and Archives Canada employees and the general public to enhance research capabilities and accessibility to Canada's historical heritage. While the project has been retired, understanding how this system operated provides insights into the application of AI technologies for cultural heritage digitization.
