传统笙的物理模型及声音合成方法
A physical model and sound synthesis method of Chinese Sheng
-
摘要: 为研究传统笙的物理结构与音色之间的关系,提出了传统笙的物理模型及声音合成方法。在簧舌振动模型的基础上,结合指孔和音窗的传输矩阵计算方法,建立了笙管的等效电路,并将二者结合,得到笙的完整物理模型。为了验证模型的有效性,设计了一套笙的实验系统,实现了对笙管发声过程中特征量的多通道同步测量。通过对比实验和模型仿真结果的时频域特点以及与音色相关的特征量,分析了模型的性能.结果表明,该模型可较好地模拟传统笙管的发声过程,合成笙样本能够较准确的反映真实笙样本的音色特征。Abstract: To study the relationship between the physical structure and the timbre of Chinese Sheng, a physical model of Sheng and its sound synthesis method are proposed. Based on the vibration model of reed tongue, the equivalent circuit of Sheng pipe is constructed by considering the transfer matrices of tone hole and tuning slot. By combining the reed and the pipe, a complete physical model of Sheng is built. An experiment system of Sheng is designed to validate the physical model, and physical quantities of sounding Sheng pipe are measured simultaneously by multi-channels. The performance of this model is analyzed by comparing temporal-spectral features and timbre descriptors of measurement and simulation results. It shows that the proposed model simulates the sounding mechanism of Sheng pipe well, and the synthesized sound is able to capture the timbre-related features of real Sheng sound in an accurate way.