Samples of Japanese emotional speech (11 words, each produced with 7 emotions: happy, angry, sad, surprised, afraid, disgusted, neutral) were elicited from two groups of speakers: Japanese native speakers and Mandarin Chinese learners of Japanese.These words were then used as stimuli in an emotion identification experiment administered to different individuals representing two groups: Japanese, Mandarin Chinese learners of Japanese.Analyses revealed two main results.First, the two listener groups found it easier to perceive emotions produced by members of the same group.Second, for speech produced by Japanese native speakers, two pairs of emotions were especially confused but not for speech produced by Mandarin Chinese learners: disgusted/angry, and sad/neutral.Acoustic and electroglottographic (EGG) analyses revealed that learners transferred emotional speech features from their native language to Japanese.This case study provides insights into emotional speech production and perception in second-language acquisition and contributes to understanding cross-cultural communication. IntroductionEmotional speech has been an important topic, not only in the field of phonetics and psychology, but also in the technology of machine learning and high-quality speech synthesis.There are numerous discussions about the differences between spontaneous emotional expression, that concern the speaker s physical or psychological states, and acted emotional expression, that can be controlled in social situations (Gu, Zhang, & Fujisaki, 2011). Fujisaki (1996) classified the information conveyed by speech into 3 types: linguistic information,
歆玥 et al. (Sat,) studied this question.