Kent R D
Waisman Center, University of Wisconsin-Madison 53705-2280.
J Voice. 1993 Jun;7(2):97-117. doi: 10.1016/s0892-1997(05)80339-x.
This paper considers the topic of vocal tract acoustics from the three perspectives: (a) the acoustic theory of speech production; (b) contemporary laboratory methods for acoustic analysis, and (c) measurement of the acoustic signal of speech. Linear source-filter theory is the standard acoustic theory of speech production and is the foundation for remarkable advances in the analysis and synthesis of speech. Digital signal processing, the dominant laboratory method for speech analysis, enables the acquisition and recording of the acoustic speech signal but also implements quantitative algorithms largely based on linear source-filter theory. Measurements of the acoustic signal reflect the acoustic theory of speech production, laboratory methods for signal analysis, and principles of experimental phonetics. Basic issues in the three domains of theory, laboratory methods, and measurement are summarized as they pertain to the interests of the voice scientist, voice clinician, and voice teacher.
（a）语音产生的声学理论；（b）当代声学分析实验室方法；（c）语音声学信号的测量。线性源-滤波器理论是语音产生的标准声学理论，也是语音分析与合成取得显著进展的基础。数字信号处理作为语音分析的主要实验室方法，不仅能够采集和记录声学语音信号，还能实现大量基于线性源-滤波器理论的定量算法。声学信号的测量反映了语音产生的声学理论、信号分析的实验室方法以及实验语音学原理。理论、实验室方法和测量这三个领域的基本问题，是围绕语音科学家、语音临床医生和语音教师的兴趣进行总结的。
