Detection

Which system produced this romanization?

Paste a source-script name and an observed Latin form. Interscript transliterates the source through every system in the family and ranks each by edit distance to your observation. The smallest distance is the most likely system.

Script family

The detector transliterates your source text through every system in the family, then ranks by Levenshtein distance between each output and your observed romanization. Distance 0 means the system produced your observed output exactly.

Why detection matters

Attribution forensics for transliteration.

You find a romanized name in an old index, a citation, or a news archive — but you don't know which romanization system was used. Interscript's detector reverse-engineers the attribution by testing every plausible system.

Provenance research

Trace which romanization system was applied to a historical record. Especially useful for archives that span multiple eras of cataloging practice.

Quality control

Verify that a romanization was applied consistently. If distance is high for every system, the source may have errors or use an undocumented convention.

Cross-system deduping

Detect when the same name surfaces under different romanizations across records — useful for entity resolution in news archives and library catalogs.