New transcription technology makes handwritten historical documents readable online
Machine-reading technology, developed with input from UCL Laws, means that millions of historical documents can now be transcribed, read and learnt from.
The lessons held in the world’s archives still have plenty to teach us, and discovering the secrets held in the handwritten ones, is now much easier thanks to a UCL project.
The Transkribus platform, developed with input from UCL Laws, has provided a model for transcribing any handwritten historical documents. The platform was developed as part of the European Research Council-funded READ project, using data from the Transcribe Bentham initiative.
Transcribe Bentham is an award-winning participatory initiative which launched in 2010 at the Bentham Project in UCL Laws. Its aim is to engage the public in the online transcription of around 100,000 pages of manuscripts written by the philosopher and reformer Jeremy Bentham.
Over 40,000 pages, or around 16 million words, have now been transcribed by volunteers, with transcription expected to be completed by 2030. The prospect that the Bentham Papers, a resource of international historical and philosophical significance, would be made digitally available was unthinkable a decade ago.
But it is the wider impact of Transkribus that makes it not just relevant to those interested in Bentham, but to any researchers or members of the public looking to access handwritten historical records.
Transcripts produced by Transcribe Bentham volunteers have been used to train the Transkribus platform to read eighteenth and nineteenth century English handwriting.
The technology is now freely available for anyone to use, and over 100,000 people have already registered for an account.
Transkribus has since been used to start translating other important historical documents, including the papers of Isambard Kingdom Brunel. Scholars who used the technology to transcribe 120 interviews of the Irish revolutionary Ernie O’Malley say it saved them ‘several years of work’.
It is hoped this project will help raise awareness of the collaborative manuscript transcription community, and encourage more academics and volunteers to collaborate on humanities research.
Find out more
Image credit: © UCL Digital Media