๐Œ๐Œ๐Œ„๐ŒEtruscan

An open-source digital corpus and computational toolkit for the study of Etruscan epigraphy. MIT / CC0 licensed.

4,728Inscriptions
45Provenances
41Pleiades Links
5Script Systems

Georeferenced Corpus

Browse 4,700+ inscriptions on an interactive map. Each entry is aligned to Pleiades and GeoNames gazetteers for interoperability with other ancient-world datasets.

Script Normalizer

Convert between five transcription systems: canonical, CIE, philological, Old Italic Unicode (U+10300), and IPA. Includes automatic source-system detection.

Inscription Classifier

Character-level neural models (CNN, Transformer) classify inscriptions by epigraphic type. Inference runs client-side via ONNX Runtime. Models published on Hugging Face.

Linked Open Data

The full corpus is exported as RDF/Turtle using LAWD and Dublin Core ontologies. A SPARQL endpoint enables cross-corpus queries within the Linked Open Data ecosystem.