Reading handwritten German memoirs

A sociology professor had images of handwritten memoirs about Sigmund Freud and wanted text he could run through translation and analysis tools. The DataSquad converted the images to text.

Two things made it hard. The memoirs were handwritten in German, and the paper had yellowed with age, which makes it harder to separate the writing from the background. The team worked through both and produced complete, accurate text.

The project shows the range of what “OCR” means in practice. The same request, turn these images into text, can be a printed page, a medieval manuscript or handwriting on aged paper, and each needs its own approach.