2021 Fiscal Year Final Research Report
A motor-commend based speech synthesis with the trigger activated by arm motions.
Project/Area Number |
19K12086
|
Research Category |
Grant-in-Aid for Scientific Research (C)
|
Allocation Type | Multi-year Fund |
Section | 一般 |
Review Section |
Basic Section 61020:Human interface and interaction-related
|
Research Institution | Tohoku Institute of Technology |
Principal Investigator |
Ito Masashi 東北工業大学, 工学部, 教授 (00436164)
|
Project Period (FY) |
2019-04-01 – 2022-03-31
|
Keywords | 音声合成 / 生体計測 / 運動 |
Outline of Final Research Achievements |
Two types of input motions were examined for synthesizing glide /w/ and /y/ with a motion-to-speech synthesizer. Using the synthesizer, subjects were required to reproduce the identical speech signals to the target one which presented beforehand. The reproduced speech parameters showed significant overshoots from the targets. The amounts of the overshoots were smaller when the synthesizer was controlled by hand motions than the case where it was controlled by arm motions. The difference became prominent in the fast transition of the speech parameters. The results indicated the mass of the body to be important factor for precious control in the synthesis.
|
Free Research Field |
音響信号処理
|
Academic Significance and Societal Importance of the Research Achievements |
Motion-To-Speech型音声合成は、失声者の代替発声装置としての応用が期待されており、誰でも簡単に使えるシステムを実現することが課題となっている。本研究では、この様なシステムの入力として用いる身体動作の動特性を定量化した点に意義があり、これまで個々の研究者が試行錯誤的に行ってきた入力動作の選定に関して、身体工学的な視点からより理に適った設計指針を提供するものであると位置づけられる。本研究で明らかになった身体部位の質量に関する知見を活用することで、より操作性の高いMotion-to-speechシステムが実現できる可能性がある。
|