Mario Cossío Olavide · PyLaia · Published January 30, 2024

14th Century Spanish Gothic Hybrid (model A)

Text Recognition

Description

Late-14th century Spanish Gothic Transitional Hybrid Cursive script (gótica cursiva híbrida de transición de fracturada a redonda), similar to Albalaes handwriting, using the semi paleographic transcription system developed by the Hispanic Seminary of Medieval Studies. Based on Biblioteca General Histórica de la Universidad de Salamanca, ms. 1958. Trained on 34394 words, 4573 lines (76 double column folios). CER on Train: 1%, CER on Validation: 6.81%. Developed for the Lucidarios Project (https://lucidarios.hypotheses.org/), as model Lucidario B (v. 0.9) by Mario Cossío Olavide (cossio@umn.edu).

Try this model

14th Century Spanish Gothic Hybrid (model A)
Use this modelOpen in Transkribus
Low error rate6.8% CER

Character Error Rate (CER) measures the percentage of characters incorrectly recognised. Lower is better. This model scored 6.8% on its validation set. As a rule of thumb, a CER below 10% is considered good for most handwritten material. This is a smaller, specialised model. It may achieve a very low CER on material similar to its training data, but could be less robust on unfamiliar handwriting or layouts.

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.

Words34,394
Lines4,573
Training Pages38
Model ID59050
Languages
Castilian