Acoustic Study Of High Vowels And Its Friction Of Changzhou Dialect

In this paper,6 high vowels(mainly [i,u,y,η,i,ч]) in Changzhou dialect and its friction properties are explored through acoustic experiment. Changzhou dialect is an important part of northern Wu piling pieces; it is adjacent with Mandarin to the north, and the speech is full of characteristics. Scholars have done some research on Changzhou dialect, but few involves vowels, while the high vowels in Changzhou dialect is particularly characteristic, you can detect the friction properties obviously. This article mainly uses the method of experimental phonetics to explore the friction properties, including analysis of formant data, drawing the vowels acoustic diagram, observing spectrogram; and the friction properties are explored using the three parameters, that are frequency domain-energy value, the resonance peak curvature and two order moments. This study first conducts a field investigation of high vowels of Changzhou dialect to obtain the first hand voice data. Then uses Emu to carry on the annotation and the extraction of the resonance peak of the experimental materials. The statistical work is conducted by R. In order to make the test results more scientific, the recording work is conducted in different bases:single character base and carrier character base; and during this way to compare the differences between single character base and natural language in the flow. A single font is divided into three structures to discuss its friction propertied respectively:CV,CVC,CGV. Among them, C represents a consonant,;V represents a vowel; and G represents [i][y][u][ч] work as medial.This paper consists of three chapters. The first chapter is introduction, mainly discusses the object of the study, the purpose of the study, the achievements of this field, and also the experimental method and program of this research. The second chapter is the general study of acoustic characteristics of high vowels in Changzhou dialect. Discussion of its acoustic properties mainly through calculating the resonance peak data, drawing the vowel chart, calculating the stable segment length. The third chapter is the most important of this article. It explores the friction properties of high vowels in Changzhou dialect. The main method is to observe the high vowel 3D spectrum and waveform, to compare the energy value of different frequency of high vowels; bases on energy value of different frequency, the resonance peak curvature and the two order moment. Using node classification of single parameter, SVM pattern classification of two parameter and dynamic three-dimensional map of three parameter, respectively, to explore the nature of friction and to make a pattern classification of high vowels.