Towards a Quantitative Analysis of Coarticulation with a Phoneme-to-Articulatory Model

TL;DR

This paper introduces a phoneme-to-articulatory model trained on EMA data to analyze coarticulation patterns over longer ranges than previous studies, providing new insights into speech production.

Contribution

It develops a model that generates realistic articulatory data for novel phoneme sequences, enabling comprehensive coarticulation analysis beyond limited phonemic contexts.

Findings

Longer-range coarticulation effects identified

Model replicates known coarticulation patterns

Potential for cross-population and cross-language studies

Abstract

Prior coarticulation studies focus mainly on limited phonemic sequences and specific articulators, providing only approximate descriptions of the temporal extent and magnitude of coarticulation. This paper is an initial attempt to comprehensively investigate coarticulation. We leverage existing Electromagnetic Articulography (EMA) datasets to develop and train a phoneme-to-articulatory (P2A) model that can generate realistic EMA for novel phoneme sequences and replicate known coarticulation patterns. We use model-generated EMA on 9K minimal word pairs to analyze coarticulation magnitude and extent up to eight phonemes from the coarticulation trigger, and compare coarticulation resistance across different consonants. Our findings align with earlier studies and suggest a longer-range coarticulation effect than previously found. This model-based approach can potentially compare coarticulation between adults and children and across languages, offering new insights into speech production.