| Project/Area Number |
18520300
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Linguistics
|
|---|
| Research Institution | Tokyo University of Agriculture and Technology |
|---|
Principal Investigator |
MISONO Yasuko Tokyo University of Agriculture and Technology, International Center, professor (00209777)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
KATO Yukari Tokyo University of Agriculture and Technology, Center for Higher Educational Development, associate professor (90376848)
|
|---|
| Project Period (FY) |
2006 – 2007
|
| Project Status |
Completed (Fiscal Year 2007)
|
| Budget Amount *help |
¥1,350,000 (Direct Cost: ¥1,200,000、Indirect Cost: ¥150,000)
Fiscal Year 2007: ¥650,000 (Direct Cost: ¥500,000、Indirect Cost: ¥150,000)
Fiscal Year 2006: ¥700,000 (Direct Cost: ¥700,000)
|
|---|
| Keywords | phonetics / spoken language / prosody / pitch pattern / intensity / prosodic information / interpretation of prosodic information / 発話意図理解 |
|---|
| Research Abstract |
The purpose of the study is to examine optimal pitch contours for "sukizyanai<negative>" and "sukizyanai<affirmative>." These two forms are grammatically different ; "sukizyanai<negatibve>" is "sukizya(adjective) + nai(auxiliary)," "sukizyanai <affirmative>" is "suki(adjective stem) + zyanai(sentence final element)."
Materials of the stimuli used in listening experiments were two of the recorded utterances, a negative and affirmative /skizyanai/, of a female professional speaker.
The FO_s of /a/ and /nai/ in /skizyanai/s were analyzed, modified by a semitone and re-synthesized using Fujisaki's model. We made six varieties of FO_s for each vowel. In this way, we made 6 x 6 x 2 = 72 stimuli.
In listening experiment A, the subjects were asked to decide whether each stimulus could be interpreted as a negative/affirmative utterance. In listening experiment B1 and B2, the subjects were asked to evaluate how good each stimulus was as a negative/affirmative utterance. The results can be considered as the subject's grading of the pitch contour of each stimuli.
Characteristics of the pitch contours which were graded highly as negative/affirmative were as follows :
1. Typical pitch contours of "sukizyanai<negative>" fall steeply from /i/ to /zya/, with a modest peak corresponding to /nai/.
2. Typical pitch contours of "sukizyanai<affirmative>" have a smoother curve.
3. When FO of vowel /a/ is high, the pitch contours tend to be evaluated as affirmative. When FO of vowels /a/ and /ail are low, the pitch contours tend to be evaluated as negative.
In the second year, we examined the effects of magnitude of the speech on the negative-affirmative decision and found that the pitch contours had far greater effects on the decision.
The results also showed that the FO_s of the all vowels are relevant to interpretation of /skizyanai/.
|
