Dionisio 2.0

Free Public AI Model for Handwritten Text Recognition with Transkribus

Dionisio 2.0

Prof. Vladimir Polomac (University of Kragujevac, Serbia) created the Dionisio 2.0. generic model for automatic recognition of Serbian Church Slavonic printed books of the 15th-17th centuries. The model was created based on the materials from various Serbian printing houses of the 15th-17th centuries (Cetinje, Venice, Goražde, Mileševa, Belgrade and Mrkša’a Church).

The percentage of incorrectly recognized characters on the validation set is 2.4%. The process of creating the model within CITLab HTR+ engine, its quantitative and qualitative performance are described in detail in a separate paper.

While the models based on the CITlabHTR+ technology are no longer available in Transkribus, the Dionisio 2.0. model has been retrained with PyLaia technology with the almost identical performances.

Model Overview

Dionisio 2.0.
Vladimir Polomac
Model ID:
15th, 16th
Serbian Church Slavonic
Cyrillic alphabet
CER on validation set:
2.40 %
Simply upload a picture and test this model

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

Dionisio 2.0. 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 Cyrillic alphabet documents into editable and searchable text.