The Transkribus Guided Project

New

We introduce the Transkribus software system that can be taught to read handwriting from images of documents and rapidly convert it into useful digital formats. This guided course provides basic training by practical immersion in this software, which requires only basic IT skills. Transkribus was developed by READ under the Horizon 2020 funding framework and is now a co-operative. It had 20,000+ users in 2019, and is becoming a standard research tool for mass transcription of archival sources. Participants will transcribe anonymised data from pre-loaded scans of forms filled out for the French national census of 1999 in Transkribus's downloadable software interface. These manual transcriptions will help train a handwritten text recognition (HTR) model to automatically transcribe many more of these forms later. In fact, the model will eventually allow the creation of one of the largest data sets ever attempted from manuscript sources. This course is a collaboration with Transkribus and Cambridge Digital Humanities. It is funded by a Cambridge Humanities Research Grant.

Sign-up is open to any current graduate student/staff at Cambridge.

Number of sessions: 2

This practical introduction course is designed to train those who are unfamiliar with Transkribus to a level where they can use it in their own research at a basic level.

Time requirements:

Two weekly video sessions on Teams: 4 - 4:30pm on 29th July and 4 - 5pm on 5th August. Task support will be provided on Teams chat between these sessions by Alexis and Oliver.

The 30 minute session on 29th July will be a welcome video session with Oliver and participants.

During the week, around 1 hour will be needed to go through pre-prepared slides containing task instructions and web links to Transkribus software. Transkribus are available by email to help with sign-up problems.

Students will spend approximately 3 hours transcribing names and addresses from digitised French census forms from the 1999 national census in Transkribus platform alone and following written instructions and links to information online. Transcriptions will be 10 forms / 5 lines of text per form.

On the 4th August, before we meet again for the final session of 5th August, participants are expected to spend 1 hour correcting the transcriptions of one other fellow participant to enable us to create the HTR model before the final session.

The work students will be managed using a spreadsheet accessible on Teams.

Upon sign-up of a place we'll add you to our CDH Learning Team channel and send instructions before the sessions.

• Apple OS or Windows PC. Sign-up for Transkribus account and download and install Transkribus software. Our pre-shared slides on the Teams channel will contain instructions and links.

• Students to self-manage work packages using a shared spreadsheet on Teams.

• Commitment to work packages. Students must commit to following the work outlined above: 5hrs course work, 1.5 hrs taught video session.

• Students must agree not to make any copies of census forms we provide.