Open AccessLexical Strata and Phonotactic Perplexity Minimization
Author(s) -
Eric Rosen
Publication year - 2021
Publication title -
proceedings of the annual meetings on phonology
Language(s) - English
Resource type - Journals
ISSN - 2377-3324
DOI - 10.3765/amp.v9i0.4918
Subject(s) - phonotactics , perplexity , computer science , grammar , artificial intelligence , natural language processing , context (archaeology) , linguistics , language model , speech recognition , phonology , history , philosophy , archaeology
We present a model of gradient phonotactics that is shown to reduce overall phoneme uncertainty in a language when the phonotactic grammar is modularized in an unsupervised fashion to create more than one sub-grammar. Our model is a recurrent neural network language model (Elman 1990), which, when applied in two separate, randomly initialized modules to a corpus of Japanese words, learns lexical subdivisions that closely correlate with two of the main lexical strata for Japanese (Yamato and Sino-Japanese) proposed by Ito and Mester (1995). We find that the gradient phonotactics learned by the model, which are based on the entire prior context of a phoneme, reveal a continuum of gradient strata membership, similar to the gradient membership proposed by Hayes (2016) for the Native vs. Latinate stratification in English.