This study examines the patterns of F0, intensity and formants in the Seoul Korean vowels preceding and following lax, aspirated, and tense consonants (/ta/, /tha/, /t'a/, /tsa/, /tsha/, /ts'a/, /sa/, /s'a/, /ata/, /atha/, /at'a/, /atsa/, /atsha/, /ats'a/, /asa/, /as'a/). The results of acoustic analysis found that the patterns of F0 and F1 in the following vowels differ among lax, aspirated, and tense consonants in both the initial and the intervocalic positions. The differences between patterns are more apparent in the initial than in the intervocalic. However, it is reasonable to believe that the patterns of F0 have no influence on consonant judgment because the angles of inclinations were subtle. On the other hand, F1 of the tense consonants differ from those in the other consonants as they rise in the first half of the following vowels. This, therefore, suggests that the patterns of F1 can influence the perception judgments of the consonants examined in this study.
The purpose of this study is to demonstrate the diachronic change in Voice Onset Time (VOT) of Korean word-initial stops associated with gender differences. Seventy-four native speakers of Seoul Korean, aged 15–59 years (born 1955–1999), were recruited for the production test. The results presently indicate three discernible stages of sound change in VOT value of stops—distinction, overlap, and merger—and females lead the change in VOT. As of 2015, Males in their 40s and 50s have a clear distinction in VOT between lenis and aspirated stops, while males in their 30s and under overlap the two stops but do not merge them. On the other hand, females in their 40s and 50s overlap the VOT between lenis and aspirated stops, whereas females in their 30s and under completely merge the two stops.
This paper studies the lengthening of Japanese speech segments taken from a picture description task. The results show a much higher occurrence of lengthening than previous studies. The degree of lengthening in each utterance was on the average 2.16 times as long as the mean duration of the preceding morae in that utterance. Almost 90% of substantial lengthening occurred at the phrase-final position of the bunsetsu. The contexts of lengthening suggest that the lengthening within bunsetsu phrases may be caused by hesitation, while bunsetsu-final lengthening may play a role in both clarifying the syntactic boundary and making speech more rhythmical.
This study examines the effect of adult speech style on face-scanning behavior in Japanese infants at six and 12 months of age. The adult speech style included three conditions: 1) Adult-directed speech (ADS) condition, 2) Infant-directed speech (IDS) condition, and 3) infant-directed singing (singing) condition. To record the infants' gaze behavior, we used an eye-tracker. Our results showed that infants in both age groups observed the mouth for a longer duration in the singing condition than in the other conditions. This main finding suggests that singing by adults may promote audio-visual speech perception and language acquisition in infants.
The purpose of this study is to investigate whether the voiced-ness of Japanese affricates and long and short vowels following the affricates, influence the length of those affricates produced by Korean learners. 46 words produced by 40 students were judged for the naturalness in terms of their length. The results revealed the voiced affricate [ʥ] preceding a short vowel /o/ and the voiceless affricate [ʨ] preceding a long vowel /oo/ were rated highly unnatural. An acoustic analysis shows the differences between the affricates judged as natural and those as unnatural ascribe to the duration of frication.
The sentence-final particle ne, if used with rising intonation, expresses seeking confirmation. However, when rising ne is connected to a sentence nominalized by no(da), it functions to share information that is inaccessible to listeners. It is interpreted as seeking confirmation only when listeners can directly access the information. Nominalization by no(da) before rising ne indicates the listener's or speaker's understanding is incomplete. In analyzing the discourse functions of ne, besides types of intonation, we should consider the existence or absence of noda.
Our daily life abounds in various characteristic voices, such as train announcements, vendor's cries, and anime dialog. Although phonetic science in general has not paid much attention to these voices, each of them seems to have its own phonation, resonance, phrasing, and intonation. Can we capture these voices “in residue”? If so, how, and to what degree? As a case study for constructing prerequisite for examining such questions, this paper deals with Japanese idioms kuchi-o togaraseru (purse one's lips) and kuchi-o yugameru (curl one's lips) and their surroundings, and clarifies how they work in daily communication by using audio-visual data.
This article qualitatively and quantitatively examines the pronunciation methods for each of the following four types of speech in Japanese: (a) objecting using childlike speech, (b) adult expressions of indebtedness, (c) directly quoting the speech of an object of disdain, and (d) “unmarked” utterance. By using video imaging and MRI movie data, we clarify the connection between these speech types and their articulatory dynamics which the idiomatic expressions kuchi-o togaraseru (purse one's lips) and kuchi-o yugameru (curl one's lips) do not cover fully.
As a case study of the general theme of this special issue—“different phonetic realizations of different social characters”, this paper explores prosodic features of two types of prototypical maid voices, which have been emerging in recent Japanese culture: “moe” and “tsun”. Two professional voice actresses read simple Japanese SOV sentences in three types of voices: Moe, tsun, and normal. Acoustic analyses show that the moe voice is characterized by higher f0 and louder voice than the normal voice, whereas the tsun voice is characterized by lower f0 and quieter voice. The current study also finds that the speakers manipulate H-tone targets more extensively than L-tone targets to differentiate different speech styles, which is compatible with some previous studies and models of intonation. In terms of its research value, the current findings may not be ground-breaking; however, an additional value of this research lies in the fact that this sort of material makes phonetic analyses more accessible to the general public as well as to students in undergraduate education. To that end, some sample sounds are made available at http://bit.ly/1WCu5DA.