音声のカオス・フラクタル性とその生成モデル

概要

論文の詳細を見る
In the field of synthesis of vocal sounds, various models composed of glottal sound source, vocal tract, and radiation into space, have been proposed. These models, however, do not consider how to involve the chaos property into them, which are not chaotic. In the present work, we proposed an oscillation model of voice synthesizer with chaotic behavior. This model utilized an exponential spring and damping in the self-oscillating model of the vocal cords. The present results show that the synthesized voices involve the chaotic property by means of the Lyapunov analysis and the analysis of surrogate data.
長岡技術科学大学の論文
2000-03-29