Bridge: Customizable Vocabulary Lists

With Bridge/Lemmatizer you can find the dictionary form (lemma) of every word in a Latin or Greek text.

1. Paste plain text to lemmatize below or upload a .txt file

Or upload a .txt file

2. Tell us what language the text is in

3. Does this text use line breaks for locations? (i.e, is it a poem, play, or song?)

4. In which format would you like your lemmata?

Hybrid (CLTK) combines all three approaches for the best accuracy. Common function words are resolved from a hardcoded table, then CLTK and Stanza are BOTH consulted for every word. When they agree, the answer is high confidence; when they disagree, the word is flagged for you to check. Recommended for most texts, and selected by default.

CLTK (Classical Language Toolkit) is a rule-based lemmatizer for ancient languages. It looks each word up in a lexicon, applies morphological rules, and falls back to the most frequent lemma. Fast and strong on the bulk of classical vocabulary, but it can leave rare or unusual forms unresolved.

Stanza uses a neural network trained on classical Latin and Greek texts. Better at rare words and unusual forms, but slower and occasionally unreliable on common function words.

Bridge uses a pre-built morpheus lexicon dictionary to quickly look up lemmas. Fast and reliable for common words, but limited in coverage for rarer forms.

5. If you download a CSV, what should it be called?

6. Click Lemmatize! to see your results on this page as flowing text with hoverable lemmas, or as a table (toggle between them at the top of the results). Or click Download CSV without waiting to skip straight to the spreadsheet, which is ready to use here.

Results