Danish-Norwegian (and German) Print 17th century

Free Public AI Model for Handwritten Text Recognition with Transkribus

Danish-Norwegian (and German) Print 17th century

Model trained on 77 Danish-Norwegian and 23 German small blackletter prints (broadside ballads – “skillingsviser”) compiled in “Peder Rafns visebog”, København 1583-1634: https://urn.nb.no/URN:NBN:no-nb_digibok_2008012413001
In the training data the different printing practices of the period are reflected. The letters i, j, u and v are transcribed diplomatic; i.e. as they appear.The training data include the ligatures β, ꜩ, œ and æ, common Latin abbreviations, and special characters such as  macrons and rotunda r (ꝛ).

About 100000 words have been trained for this model, and the CER for the validation set is 1.53%.

Model Overview

Name:
NorFraktur_1600_v0.3
Creator:
National Library of Norway
Model ID:
38253
Century:
17th
Languages:
Danish, German, Norwegian
Script:
Latin alphabet
Engine:
HTR+
Material:
Print
CER on validation set:
1.53 %
Simply upload a picture and test this model

By uploading an image, you accept our terms and conditions and our privacy policy

NorFraktur_1600_v0.3 is freely available to everyone

Get started with Transkribus and use it for your own Material
You can use this model to automatically transcribe Print documents with Handwritten Text Recgnition in Transkribus. This model can be used in the Transkribus Expert Client as well as in Transkribus Lite.
This AI model was trained to automatically convert text from images of historical Latin alphabet documents into editable and searchable text.