margret.jona.gisladottir · PyLaia · Published December 13, 2025
Icelandic 17th Century (experimental)
Text Recognition
Description
Experimental model for 17th century Icelandic handwriting. The baseline model for it is: Icelandic late 18th century (experimental). The materials used for this model are Skarðsárannáll Lbs 40 (1641–1642) (to p. 26, or 23 pages) and the priest's book of Skálholt diocese from the years 1639–1674 (pages 1–109), Bps. A III-1, written by different scribes. Abbreviations are mostly suspended. Only vowels with acute accent on Óó (ö) and Öö (ó) are kept; all other vowels are transcribed without acute accents.
Try this model
Use this modelOpen in Transkribus
Low error rate5.91% CER
Character Error Rate (CER) measures the percentage of characters incorrectly recognised. Lower is better. This model scored 5.91% on its validation set. As a rule of thumb, a CER below 10% is considered good for most handwritten material.
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.
Words33,824
Lines4,469
Training Pages69
Model ID451505