Relative Contributions of Spectral and Temporal Cues to Korean Phoneme Recognition

Bong Jik Kim; Son-A Chang; Jing Yang; Seung-Ha Oh; Li Xu

doi:10.1371/journal.pone.0131807

Relative Contributions of Spectral and Temporal Cues to Korean Phoneme Recognition

PLoS One. 2015 Jul 10;10(7):e0131807. doi: 10.1371/journal.pone.0131807. eCollection 2015.

Authors

Bong Jik Kim¹, Son-A Chang², Jing Yang³, Seung-Ha Oh⁴, Li Xu⁵

Affiliations

¹ Department of Otorhinolaryngology, Seoul National University College of Medicine, Seoul, Korea.
² School of Speech Language Therapy and Aural Rehabilitation, Woosong University, Daejeon, Korea.
³ Communication Sciences and Disorders, University of Central Arkansas, Conway, Arkansas, 72035, United States of America.
⁴ Department of Otorhinolaryngology, Seoul National University College of Medicine, Seoul, Korea; Sensory Organ Research Institute, Seoul National University Medical Research Center, Seoul, Korea.
⁵ Communication Sciences and Disorders, Ohio University, Athens, Ohio, 45701, United States of America.

Abstract

This study was aimed to evaluate the relative contributions of spectral and temporal information to Korean phoneme recognition and to compare them with those to English phoneme recognition. Eleven normal-hearing Korean-speaking listeners participated in the study. Korean phonemes, including 18 consonants in a /Ca/ format and 17 vowels in a /hVd/ format, were processed through a noise vocoder. The spectral information was controlled by varying the number of channels (1, 2, 3, 4, 6, 8, 12, and 16) whereas the temporal information was controlled by varying the lowpass cutoff frequency of the envelope extractor (1 to 512 Hz in octave steps). A total of 80 vocoder conditions (8 numbers of channels × 10 lowpass cutoff frequencies) were presented to listeners for phoneme recognition. While vowel recognition depended on the spectral cues predominantly, a tradeoff between the spectral and temporal information was evident for consonant recognition. The overall consonant recognition was dramatically lower than that of English consonant recognition under similar vocoder conditions. The complexity of the Korean consonant repertoire, the three-way distinction of stops in particular, hinders recognition of vocoder-processed phonemes.

Publication types

Research Support, Non-U.S. Gov't

MeSH terms

Adult
Cochlear Implants
Cues
Female
Humans
Male
Pattern Recognition, Automated
Phonetics
Republic of Korea
Signal Processing, Computer-Assisted
Speech Perception*
Young Adult

Grants and funding

This research was supported by the Brain Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT & Future Planning (No. 2014043923). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.