PLD2flex: Establishing the Phonological Levenshtein Distance for Pairs or Groups of (Pseudo)Words

Abstract

Establishing the phonological Levenshtein distance (PLD) of words and pseudowords is useful for various psycholinguistic research applications, such as generating stimuli for experiments on language processing or analysing the PLD between erroneous and intended utterances. PLD2flex is a tool for establishing the PLD of (pseudo)words in pairwise, one-to-many and many-to-many comparisons of orthographic word forms, including one-to-many comparisons of (pseudo)word forms with databases of words generated from corpora. Specifically, one-to-many comparisons can be used to determine the average distance of a word to its 20 closest neighbours (PLD20), in analogy to the OLD20-measure for written words. PLD2flex makes use of the BAS Web Services API and several third-party python libraries. PLD2flex is freely available at https://github.com/FelixTheodor/PLD2flex.

Keywords

• levenshtein distance
• phonological distance
• multilingual
• lexical databases
• cross comparison
• psycholinguistics
• phonology
• orthography