| Project/Area Number |
15K12059
|
|---|
| Research Category |
Grant-in-Aid for Challenging Exploratory Research
|
| Allocation Type | Multi-year Fund |
|---|
| Research Field |
Perceptual information processing
|
|---|
| Research Institution | The University of Tokyo |
|---|
Principal Investigator |
Minematsu Nobuaki 東京大学, 大学院工学系研究科(工学部), 教授 (90273333)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
齋藤 大輔 東京大学, 大学院工学系研究科(工学部), 講師 (40615150)
|
|---|
| Research Collaborator |
UCHIDA Hidetsugu
|
|---|
| Project Period (FY) |
2015-04-01 – 2017-03-31
|
| Project Status |
Completed (Fiscal Year 2016)
|
| Budget Amount *help |
¥3,640,000 (Direct Cost: ¥2,800,000、Indirect Cost: ¥840,000)
Fiscal Year 2016: ¥1,820,000 (Direct Cost: ¥1,400,000、Indirect Cost: ¥420,000)
Fiscal Year 2015: ¥1,820,000 (Direct Cost: ¥1,400,000、Indirect Cost: ¥420,000)
|
|---|
| Keywords | 音声・調音推定 / EMA / パラレルコーパス / 話者正規化 / 音声の構造的表象 / 外国語学習 / 調音運動計測 / 調音・音響マッピング / 統計的声質変換 |
|---|
| Outline of Final Research Achievements |
In this study, by integrating three techniques, articulatory measurement, acoustic measurement, and statistical media conversion, novel techniques were built, which can convert speech signals into the articulatory movement pattern that is assumed to be conducted to generate the speech signals. Further, a corpus was developed, which is required to realize the above technique. As a result of this study, 1) by using EMA, a parallel corpus between acoustic measurements and articulatory measurements was built from a single Japanese-Chinese bilingual speaker. 2) the performance of acoustic-to-articulatory mapping was improved by introducing speaker normalization techniques. 3) by using the structural representation of speech, a novel technique was built to predict the articulatory movement of a phoneme, which is impossible for the speaker to generate correctly. These results were presented at top conferences of speech communication and a journal paper was published.
|
