This dataset comprises 74 digitised images (TIFF files) drawn from a selection of early printed Bengali books (1713-1914) digitised through the Two Centuries of Indian Print project (https://www.bl.uk/projects/two-centuries-of-indian-print). Also contained are ground truth transcriptions (XML) for each page that can be used for training optical character recognition software on historical Bengali printed text. The folder contains images and ground truth used to train Transkribus. The images are out of copyright and the transcriptions are also public domain.