Lexical strata and phonotactic perplexity minimization
- Eric R Rosen (Johns Hopkins University)
Abstract
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).
Keywords: lexical strata, gradient phonotactics, neural models
How to Cite:
Rosen, E. R., (2021) “Lexical strata and phonotactic perplexity minimization”, Society for Computation in Linguistics 4(1), 415-419. doi: https://doi.org/10.7275/rx52-t759
Downloads:
Download PDF