Acoustic Features of English Fricatives Produced by Chinese EFL Learners at Different English Proficiency Levels

This study acoustically analyzes the production of English fricatives (/f/, /v/, /θ/, /ð/, /s/, /z/, /∫/, /ʒ/) by ten native speakers from the United States and 40 Chinese EFL learners (of various English proficiency levels). Six acoustic parameters proved to be capable of distinguishing places of articulation, voicing and vowel contexts from previous studies are considered: spectral mean (M1), variance (M2), skewness (M3), kurtosis (M4), relative amplitude and normalized duration. The statistical results reveal that: Firstly, the six acoustic parameters work differently in distinguishing places of articulation, voicing, and vowel contexts between native speakers and advanced learners and beginners. The effect of vowel context becomes weaker with the improvement of L2 proficiency. Secondly, Chinese EFL learners demonstrate significant differences in acoustic features with native speakers in English fricative production, which indicates that Chinese EFL learners face challenges in producing native-like English fricatives. Besides, advanced learners are found to demonstrate a performance more similar to native speakers (less significant differences in acoustic characteristics with native speakers) than beginners. Lastly, segmental substitutions errors of /v/-/w/ and /ʒ/-/r/ are observed in Chinese EFL learners. (Wonkwang University)