transkribering.samla · PyLaia · Published September 29, 2025
Nordic typewriter 1900-1950
Text Recognition
Description
Based on Norwegian, Swedish and Danish typed material. It should be able to recognize æ – ø – ö – å – ä, and distinguish between them.
The texts used for the model are catalogue index cards, letters, folklore records from the digital tradition archive SAMLA (samla.no).
SAMLA is a collaboration between The Ethno-Folkloristic Archive at the University of Bergen (owner), The Norwegian Folklore Archives at the University of Oslo, The Norwegian Ethnological Research at the Norwegian Folk Museum.
Developed by Therese Foldvik (University of Oslo).
Try this model
Use this modelOpen in Transkribus
Very low error rate1.22% CER
Character Error Rate (CER) measures the percentage of characters incorrectly recognised. Lower is better. This model scored 1.22% on its validation set. As a rule of thumb, a CER below 10% is considered good for most handwritten material. This is a larger model trained on diverse material, which generally makes it more robust across different handwriting styles. That said, larger training sets also make it harder to push the CER down further.
Measured on the model's own validation data. Results on your documents may differ depending on handwriting style, document condition, language, and how closely your material resembles the training data.
Words73,001
Lines8,709
Training Pages598
Model ID408777