Acoustic and Perceptual Differences of Aegyo Speaking Style Across Gender in Seoul Korean

Abstract

Aegyo is a culturally salient speaking style in Korea, often described as a baby talk-like register used by young adults to convey affection or cuteness. Yet, its acoustic profile and perception across genders remain understudied. This study investigates the acoustic and perceptual characteristics of aegyo through a production study and a perception study. In the production study, 12 native Seoul Korean speakers (six females, six males) produced sentences in both aegyo and non-aegyo styles. Acoustic analyses revealed that aegyo is characterized by significantly longer vowel durations, slower speech rate, and higher mean and maximum F0, along with greater variability in F0 and vowel duration: F0 range was significant only for male speakers. In addition, “hyper-score” analyses showed that male speakers exhibited more increases in mean and maximum F0 compared with female speakers. In the perception study, 49 Korean listeners (25 females, 24 males) judged whether the stimuli were produced in aegyo. Results showed a significant interaction between sensitivity and bias: listeners were less accurate but more prone to label the stimuli as aegyo when the speaker was male, whereas they were more accurate and more conservative when the speaker was female. These findings suggest that listeners interpret the same speaking style differently depending on speaker gender. Overall, our results support the Speaker Design model by evidencing that speakers systematically shift their vocal behavior to construct social identity, while also showing that listener interpretation of such shifts may vary by speaker gender.