No abstract

This study examines how ethnophonetic sounds are perceived in three different language/cultural groups. Specifically, Japanese, Chinese and American listeners were asked to listen to samples of voices of Japanese cake-selling street voices, and to rate which voice was the "best". The results indicate Japanese listeners are quite sensitive to what voice is best as a seller of fashionable Western cakes, and that this voice is different from sellers in less fashionable stores. The non-Japanese listeners rated the experienced Japanese cake-street seller voice considerably lower than did the Japanese listeners; moreover, Chinese and American listeners' differed on which street-seller voice they preferred. Tentative analysis suggests that Chinese listeners preferred a street selling voice with a higher F0, one that sounds like the moe anime voice, while American listeners preferred the voice with a more dynamic range of F0. Japanese listeners, on the other hand, preferred the voice that sounded "more elegant"-one with a touch of twang and some breathiness, a voice quality that is often perceived as being nasal (hana ni kakatta koe). An interesting question to be explored in the future is why the same voice is interpreted differently in different cultures. .

Voice quality differences [1] can convey different attitudes and emotions [2], with speakers of different languages showing different sensitivities to voice qualities, e.g., [3,4,5]. It remains to be explored, however, precisely which acoustic properties are perceptually associated with what emotional meanings, and whether such perceptual mappings hold universally or differ across languages. This paper offers a first step addressing these issues. Building upon the previous findings that speakers of different languages demonstrate different sensitivities to voice quality differences, the study examines particularly how the perceptions of arousal and valence are affected by different voice qualities. The current experiment reveals that speakers of the three language groups share similar ratings of arousal in association with breathy voices. Yet the valence ratings vary among the groups: Japanese and Mandarin listeners rate voices with high F0 and small OQ with positivity, whereas Brazilian Portuguese rate voices with low F0 and larger OQ with positivity. The findings of this study have applications for second language teaching, and carry over to the worlds of business, politics, and advertisement; in general, this type of research may have a potential to be useful for improving communication in crosscultural inter-personal relationships.

Objectives: To provide detailed descriptions of contraction-induced morphometric changes in the extravelar segments of the levator veli palatini (LVP) muscle using 3-dimensional (3-D) magnetic resonance imaging (MRI). Design: Three-dimensional MRI data were acquired at rest and during “silent /i/” from 4 singers. During silent /i/, participants voluntarily sustained velar elevation while breathing orally for the entire scan time. Focusing on the extravelar segments, LVP length, angle of the muscle origin, and cross-sectional area (CSA), measurements were obtained and compared between tasks. Results: Three of the 4 participants exhibited the expected patterns of change following concentric contraction of the LVP muscle. Consistent changes from the resting to the contracted state included reductions in LVP length by 13.5% and angle of the muscle origin by 9.8%, as well as increases in CSAs by 22.1%, on average. Conclusions: This study presented high-resolution data of the LVP muscle behavior with the first in vivo 3-D measurements of the contracted LVP muscle, which can be useful for the validation of computational models that aim at describing biomechanical properties of the LVP muscle in future research. The active behavior of the extravelar LVP muscle also provides some insight on optimal LVP muscle geometry to consider during cleft palate repair.

No abstract