Morphological Inflection with Phonological Features

TL;DR

This paper investigates how incorporating phonological features into neural morphological models affects performance, finding that models can infer phonological patterns from character distributions without explicit phoneme representations.

Contribution

It introduces two methods for integrating phonological features into morphological models and demonstrates their effectiveness across multiple languages.

Findings

Methods yield comparable or minor improvements over baseline

Models can infer phonological patterns from character data

Explicit phoneme features are not always necessary for good performance

Abstract

Recent years have brought great advances into solving morphological tasks, mostly due to powerful neural models applied to various tasks as (re)inflection and analysis. Yet, such morphological tasks cannot be considered solved, especially when little training data is available or when generalizing to previously unseen lemmas. This work explores effects on performance obtained through various ways in which morphological models get access to subcharacter phonological features that are the targets of morphological processes. We design two methods to achieve this goal: one that leaves models as is but manipulates the data to include features instead of characters, and another that manipulates models to take phonological features into account when building representations for phonemes. We elicit phonemic data from standard graphemic data using language-specific grammars for languages with shallow grapheme-to-phoneme mapping, and we experiment with two reinflection models over eight languages. Our results show that our methods yield comparable results to the grapheme-based baseline overall, with minor improvements in some of the languages. All in all, we conclude that patterns in character distributions are likely to allow models to infer the underlying phonological characteristics, even when phonemes are not explicitly represented.