Effects of noise, reverberation and foreign accent on native and non-native listeners' performance of English speech comprehension

Z. Ellen Peng, Lily M. Wang

Research output: Contribution to journalArticlepeer-review

32 Scopus citations

Abstract

A large number of non-native English speakers may be found in American classrooms, both as listeners and talkers. Little is known about how this population comprehends speech in realistic adverse acoustical conditions. A study was conducted to investigate the effects of background noise level (BNL), reverberation time (RT), and talker foreign accent on native and non-native listeners' speech comprehension, while controlling for English language abilities. A total of 115 adult listeners completed comprehension tasks under 15 acoustic conditions: three BNLs (RC-30, RC-40, and RC-50) and five RTs (from 0.4 to 1.2 s). Fifty-six listeners were tested with speech from native English-speaking talkers and 59 with native Mandarin-Chinese-speaking talkers. Results show that, while higher BNLs were generally more detrimental to listeners with lower English proficiency, all listeners experienced significant comprehension deficits above RC-40 with native English talkers. This limit was lower (i.e., above RC-30), however, with Chinese talkers. For reverberation, non-native listeners as a group performed best with RT up to 0.6 s, while native listeners performed equally well up to 1.2 s. A matched foreign accent benefit has also been identified, where the negative impact of higher reverberation does not exist for non-native listeners who share the talker's native language.

Original languageEnglish (US)
Pages (from-to)2772-2783
Number of pages12
JournalJournal of the Acoustical Society of America
Volume139
Issue number5
DOIs
StatePublished - May 1 2016

ASJC Scopus subject areas

  • Arts and Humanities (miscellaneous)
  • Acoustics and Ultrasonics

Fingerprint

Dive into the research topics of 'Effects of noise, reverberation and foreign accent on native and non-native listeners' performance of English speech comprehension'. Together they form a unique fingerprint.

Cite this