| Project/Area Number |
11480086
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (B)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
情報システム学(含情報図書館学)
|
|---|
| Research Institution | Kyoto Institute of Technology |
|---|
Principal Investigator |
NIIMI Yasuhisa Kyoto Institute of Technology, Faculty of Eng. and Design, Professor, 工芸学部, 教授 (00026030)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
NISHIMOTO Takuya Kyoto Institute of Technology, Faculty of Eng. and Design, Assistant, 工芸学部, 助手 (80283696)
ARAKI Masahiro Kyoto Institute of Technology, Faculty of Eng. and Design, Associate Professor, 工芸学部, 助教授 (50252490)
|
|---|
| Project Period (FY) |
1999 – 2001
|
| Project Status |
Completed (Fiscal Year 2001)
|
| Budget Amount *help |
¥9,600,000 (Direct Cost: ¥9,600,000)
Fiscal Year 2001: ¥1,400,000 (Direct Cost: ¥1,400,000)
Fiscal Year 2000: ¥4,200,000 (Direct Cost: ¥4,200,000)
Fiscal Year 1999: ¥4,000,000 (Direct Cost: ¥4,000,000)
|
|---|
| Keywords | modeling of atmosphere / communicative situation / gesture agent / recogniton of emotional speech / synthesis of emotional speech / 感情音声に合成 / 擬人化エージェント / 非同期音声会議 / 音声中の雰囲気 / ニュー波 / 感情音声の合成 / ジェスチャー認識 |
|---|
| Research Abstract |
Some atmosphere is created in situations where multiple persons communicate each other. The atmosphere of such situations is quite important in activating the discussion or the meeting. What an atmosphere is created depends on (1) what is discussed there, (2) social positions of the participants, their roles therein, and their characteristics, and (3) their facial expressions, their tones of voice, and their physical behaviors. In this project we mainly concerned the third of these factors and studied the following three points.
1. Analysis of a conversation atmosphere
Situations where two persons talked spontaneously were recorded with a video camera. The timing of gestures and the intervals of utterances were annotated for each person, and analyzed on what differences in these factors were observed between where dialogs were activated and where dialogs were not activated. Utterances of two persons overlapped each other more frequently in the former situations than in the latter, and in the intervals of the overlaps directions of a glance frequently changed in the former while movements of bodies were frequently observed in the latter.
2. Development of a gesture agent
Physical behaviors of subjects who were asked to make a gesture of some psychological states such as "good humor", "convinced", and "interested" and their counterparts were observed with a simple motion-capture. The analysis of these date proved that combinations of movements of a head, a shoulder and hands could distinguish among these psychological states. A graphical gesture agent was designed based on this result.
3. Recognition and synthesis of emotional speech
Emotion is important to create a conversation atmosphere. So a prototype system was constructed to recognize emotions such as anger, fear, joy, and sadness included in speech and a speech synthesis system was built to produce emotional speech by which three emotions, anger, joy, and sadness can be synthesized.
|
