We recorded English speech produced by Japanese elementary school children from the 4^th grade to the 6^th grade. The formant frequencies and segmental duration of 9 English monophthongs were measured on the speech of 47 children. The analysis showed that spectral variance of each monophthong decreased over time. We focused on the analysis of three vowel groups, /ᴧ/-/æ/-/ᵅ/, /ᶦ/-/ᵢ/ and /ᶷ/-/ᵘ/. Although the children had difficulties in differentiating lax and tense vowels acoustically, a significant difference was observed in F1 frequency between /æ/ and the other two, /ᴧ/ and /ᵅ/. Meanwhile, they differentiated /ᶦ/-/ᵢ/ and /ᶷ/-/ᵘ/ with segmental duration.

This study examines cross-cultural differences for the perception by Japanese, French and American judges of audio-visual recordings of a short Japanese utterance produced with nine different social affective expressions. The listeners’ task was to create an audio-visual expression that fit a given expressive label, by matching one of the nine video with one of the nine audio stimuli. L1 judges showed a higher rate of correct matching than non-L1 judges; confusions were within semantic categories. Non-L1 judges showed matching patterns similar to L1 ones, with modality-specific differences especially for culturally-related expressions like Japanese politeness and the occidental expression of seduction.

This paper focuses on a phonological research of the phenomenon of the loss of accent in standard Tokyo Japanese. The aim of this paper is to examine which of the accented patterns (initial, final, penultimate, antepenultimate or pre antepenultimate patterns) is most likely to be replaced by the unaccented pattern in case of wago (Japanese origin) and kango (Chinese origin). The main findings are that: a) the final and penultimate accent types are the accent types which are the most replaced by the unaccented pattern, b) the antepenultimate pattern is the pattern which is the most resistant to accent lost.

The current paper shows three case studies of sound symbolism in fictional names featured in popular Japanese computer games: Digital Monster, Monster Hunter, and Final Fantasy. First, we show positive correlations between evolutionary stages and the number of voiced obstruents and the number of morae in Digital Monster. This replicates the results of the sound-symbolic analysis of Pokémon names conducted by Kawahara et al. (2018a). Second, we show that only the effect of the morae is positively associated with the size (height) of the monsters in Monster Hunter. Third, we reveal a positive correlation between the spell levels in Final Fantasy and the number of voiced obstruents, but not between the spell levels and the number of morae.

This paper presents a design for a phonetics course specifically tailored to Japanese preservice teachers. It suggests using General American as a production model, while presenting diverse varieties for reception. Critically, such a course must impart three components: declarative knowledge of phonetics, procedural knowledge of pronunciation teaching, and sufficient pronunciation and listening training. Rather than covering every pronunciation detail, prioritizing features that contribute to intelligibility is necessary. Through this course, preservice teachers are expected to become equipped with knowledge and techniques for teaching English pronunciation and to develop intelligible pronunciation to serve as good models for their students.

In this paper, we illustrate the difficulties of teaching Japanese pitch accent through a detailed questionnaire survey of Japanese language teachers and propose improvements to phonetic education. Accent is important in communication for Japanese language learners, and the need is increasing due to changes in social circumstances. The questionnaire survey conducted in this study suggested that Japanese language teachers’ phonetic education should emphasize improvement of the ability to assess the pitch of one’s own voice and understanding of the difference between model and actual pronunciation.