Compositional Phoneme Approximation for L1-Grounded L2 Pronunciation Training

TL;DR

This paper introduces a compositional phoneme approximation method that improves second language pronunciation training by using native phoneme sequences, leading to more native-like speech with minimal training.

Contribution

It presents a novel feature-based CPA technique for L1-grounded pronunciation training, enhancing phoneme recognition and speech naturalness in L2 learners.

Findings

76% in-box formant rate in acoustic analysis

17.6% relative improvement in phoneme recognition accuracy

Over 80% of speech rated as more native-like

Abstract

Learners of a second language (L2) often map non-native phonemes to similar native-language (L1) phonemes, making conventional L2-focused training slow and effortful. To address this, we propose an L1-grounded pronunciation training method based on compositional phoneme approximation (CPA), a feature-based representation technique that approximates L2 sounds with sequences of L1 phonemes. Evaluations with 20 Korean non-native English speakers show that CPA-based training achieves a 76% in-box formant rate in acoustic analysis, 17.6% relative improvement in phoneme recognition accuracy, and over 80% of speech being rated as more native-like, with minimal training. Project page: https://gsanpark.github.io/CPA-Pronunciation.