tranScriptorium : A Guide for the Humanist

The aim of the tranScriptorium project is to produce software applications that will ‘read’ historical handwritten documents and will produce transcripts for these documents. There are many millions of pages of handwritten documents, the basis for much research in the arts and humanities, and yet the vast majority of this material—including manuscripts that have been digitized—remains inaccessible, except to the individual scholar who has the time and resources to visit the archive where the material is deposited and the palaeographic skills required to survey and read it. The creation of a process that will allow such material to be machine-read will have immensely beneficial consequences for scholarship and research. The vision behind tranScriptorium is that, beginning with digital images of handwritten documents, the software application will generate a typewritten transcription that not only reproduces the words themselves but also their layout on the page, and is fully searchable. The following account explains, in non-technical language, the various steps that are being taken in the tranScriptorium project order to turn this vision into reality.