Compositional Phoneme Approximation for L1-Grounded L2 Pronunciation Training

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Park, Jisang, Kim, Minu, Hong, DaYoung, Lee, Jongha
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866917069887373312
author Park, Jisang
Kim, Minu
Hong, DaYoung
Lee, Jongha
author_facet Park, Jisang
Kim, Minu
Hong, DaYoung
Lee, Jongha
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2411_10927
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Compositional Phoneme Approximation for L1-Grounded L2 Pronunciation Training
Park, Jisang
Kim, Minu
Hong, DaYoung
Lee, Jongha
Computation and Language
Sound
Audio and Speech Processing
H.5.5
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.
title Compositional Phoneme Approximation for L1-Grounded L2 Pronunciation Training
topic Computation and Language
Sound
Audio and Speech Processing
H.5.5
url https://arxiv.org/abs/2411.10927