Russian print of the 18 c. (V. Okorokov’s Printing House)

Free Public AI Model for Handwritten Text Recognition with Transkribus

Russian print of the 18 c. (V. Okorokov’s Printing House)

The model «Russian print of the 18 c. (V. Okorokov’s Printing House)» was created during the course of the project «Plants and people in the Russian Empire of the 18th century: knowledge and practice distribution among social classes» at the European University at St. Petersburg. At the first stage, several scientific treatises published at the V. Okorokov’s Printing House (at the Moscow State University) were chosen as a starting point for transcribing. The books were printed in Russian, but some scientific notions and terms were supplied with Latin equivalents.

The model also shows good results on the other printed Russian texts of the 18th c., and can be used as a base model to speed up the creation of a general model for Russian print of the 18th c., including every specific printing house.

Training sources are books, scanned by Google Books (include more than 82,000 words):  

https://www.google.ru/books/edition/Георгия_Готлиба_Рихте/_7ZiAAAAcAAJ?hl=ru&gbpv=0

https://www.google.ru/books/edition/О_дѣйствиях_и_силѣ_шал/v65iAAAAcAAJ?hl=ru&gbpv=0

Model Overview

Name:
Russian print of the 18 c. (V. Okorokov's Printing House)
Creator:
Kira Kovalenko
Model ID:
44358
Century:
18th
Languages:
Latin, Russian
Script:
Latin alphabet, Cyrillic alphabet
Engine:
PyLaia
Material:
Print
CER on validation set:
0.60 %
Simply upload a picture and test this model

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

Russian print of the 18 c. (V. Okorokov's Printing House) 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, Cyrillic alphabet documents into editable and searchable text.