thomas.zangl · PyLaia · Published May 30, 2024

Latin Humanisitic manuscript 17th century

Text Recognition

Description

This model is trained with a training data set of 60 pages of a manuscript ("Theosophia Aegyptiorum") by the alchemist and physician Michael Maier (1568-1622) from the early 17th century - presumably from 1611. The manuscript is written in Latin language and was written by a scribe. A model trained on the basis of three of Maier's concept writings (also manuscripts from the early 17th century) was used as the base model for training the model. Within the model the logical negation sign ("¬") was used as a word separator. Abbreviations - for example for "que" or "us" - were not broken up and transcribed as they appear in the text. Therefore Unicode characters have been used (ꝯ, ꝗ, ᵽ, ḡ). The same applies to any special characters (ÿ, ȳ, â, á, ê, é, ô, ó) and Ligatures have also been retained (Æ, æ). The transcription is therefore based on the original and is not orientated towards simplified readability.

Try this model

Use this modelOpen in Transkribus
Very low error rate4.42% CER

Character Error Rate (CER) measures the percentage of characters incorrectly recognised. Lower is better. This model scored 4.42% 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.

Words11,511
Lines1,555
Training Pages54
Model ID92753
Languages
Latin
Centuries
16th c.17th c.