Acoustic and Perceptual Similarities Between English and Korean Sibilants

Abstract. Foreign accent has been assumed to be closely related to the degree of articulatory, acoustic and perceptual similarity between L1 and L2 sounds. This study examined cross-language acoustic and perceptual similarities between Korean and English sibilant fricatives: Korean [—tense] /s/ and [+tense] /s*/ vs. English alveolar /s/ and palato-alveolar /∫/. To determine acoustic similarity, two parameters were measured: duration and spectral peak frequency. A Same-Different (AX) discrimination task investigated listeners' perceived similarity judgments between pairs of sibilants. In most cases, the acoustic characterizations led to correct predictions about differences in listeners' perceptions. However, results showed several disparities between acoustic similarity and perceived similarity. These cases necessarily involve acoustic dimensions other than the two measured here; probable candidates are voice quality on a following vowel, and lip rounding, with its spectral lowering effects. Cases of mismatch between acoustic and perceptual characterizations are fruitful areas for examining additional acoustic characteristics that may be responsible for listeners' ability to distinguish sounds. Acoustic and perceptual characterizations in tandem provide the best method of establishing areas of difference between the sounds of different languages, and in turn of establishing ways to teach L2 sounds to learners.