Skip to content

Pricing

Log In Get Started

Search...

Pricing

Log In Get Started

Start typing to search across models, sites, and blog posts...

HomeDatasetsModel training dataLontara v.0.08

Lontara v.0.08

Model training dataView only

View in Transkribus View the model

Access

The model owner disclosed this training data for viewing only. It is not available for reuse.

Ground-truth pages: 24
Words: 2,558
Languages: Buginese, Makasar
Model CER: 4.72%

Example page

Example page from the Lontara v.0.08 training data

Products

Transkribus
Sites
API
Scanning
Pricing

Features

Text Recognition
Field Recognition
Table Recognition
Information Extraction
Editor
Collaboration
Tagging

Solutions

Archives & Institutions
Libraries
Research & Academia
Genealogy
Developers & IT
Digitization

Explore

Models
Datasets
Sites
Success Stories
Blog
Events
Help Center
Scholarship

Transkribus is used for

Cursive Handwriting
Kurrent Script
Sütterlin Script
Council Minutes
Old German Script
Handwriting to Text
Image to Text
Handwriting OCR

Cooperative

About READ-COOP
Members
Join the Cooperative
Press & Media
Statutes
Contact Us

Legal

Privacy Policy
Terms & Conditions
Subscription Terms

Join 500,000+ users who trust Transkribus for handwritten text recognition.

Create a free account

European Cooperative

GDPR Compliant

500,000+ Users

© 2026 READ-COOP SCE. All rights reserved.

Terms of Service Privacy Policy Imprint