元音声学空间的个体特性研究

发布时间：2018-06-08 23:42

本文选题：数据采集与分析 + 元音　；参考：《电子科技大学》2014年硕士论文

【摘要】：元音声学空间的研究是声学鉴别领域的重要研究课题。本文作者的主要工作在于实践软件工程的思想,通过设计并实现数据采集系统和数据分析系统,达到对元音声学空间的研究和分析。该项研究对元音的分析有重要的应用价值和实际意义。本文首先阐述了国内外在语音级别方面的发展和应用现状、以及数据采集系统和数据分析软件。根据软件工程的各个阶段,详细介绍了数据采集系统从需求分析、设计、实现及功能介绍等阶段。本文还介绍了Praat软件在语音研究中的采集分析与标注等功能,其还可以进行多种处理任务,包括各种变换及滤波。最后其以文字报表或语言图谱的方式为结果展示给人们,作为这两种结果模式不仅能够制作成位图和矢量图,同时在根据人们的需要还可以连接到计算机显示或保存在计算机上,以便于供其他学者采用。同时,本文还简单的介绍了最大似然率理论和空间共振峰理论,可以帮助人们更好的理解本课题的研究过程。本论文的主要工作包括了设计并实现数据采集系统和数据分析软件。利用数据采集系统录取北京话、上海话和广州话的音频文件,数据分析软件使用管理人员筛选出的有效音频文件进行分析,然后利用空间共振峰理论建立空间坐标,分别对北京话、上海话和广州话进行分析,同时利用得到的数据进行折线图的分析。研究表明在三维空间相较于二维空间更容易分辨出各个元音之间的差别,包括共振峰的大小、空间位置的区别,同时也证明了,在详细的分析和正确的试验情况下,三维空间的共振峰更有利于作为语音鉴定的有力证据。三维立体空间在元音的相对关系和分布在空间的位置上的分布更具有系统性、完整性和科学性。
[Abstract]:The research of vowel acoustic space is an important research topic in the field of acoustic identification. The main work of the author is to practice the idea of software engineering. By designing and implementing the data acquisition system and data analysis system, the research and analysis of vowel acoustic space can be achieved. This research has important application value and practical significance to the analysis of vowels. In this paper, the development and application of voice level at home and abroad, data acquisition system and data analysis software are introduced. According to each stage of software engineering, the data acquisition system is introduced in detail, including requirement analysis, design, realization and function introduction. This paper also introduces the functions of Praat software in speech research, such as acquisition, analysis and annotation. It can also perform a variety of processing tasks, including all kinds of transformation and filtering. Finally, it is shown to people in the form of text reports or language atlas. As a result of these two models, not only can they be made into bitmaps and vector images, At the same time, according to the needs of people can be connected to the computer display or stored on the computer, for other scholars to use. At the same time, this paper also briefly introduces the maximum likelihood rate theory and the spatial resonance peak theory, which can help people better understand the research process of this subject. The main work of this paper includes the design and implementation of data acquisition system and data analysis software. The audio files of Beijing dialect, Shanghai dialect and Guangzhou dialect are taken from the data acquisition system, and the data analysis software is analyzed using the valid audio files selected by the managers, and then the spatial coordinates are established by using the theory of spatial resonance peak. The Beijing dialect, Shanghai dialect and Guangzhou dialect are analyzed respectively, and the broken line map is analyzed by using the obtained data. The study shows that it is easier to distinguish the differences between vowels in three-dimensional space than in two-dimensional space, including the size of the resonance peak and the spatial location. It also proves that, in the case of detailed analysis and correct experiments, The resonance peak in three-dimensional space is more favorable for the identification of speech. The relative relation of three dimensional space in vowels and the distribution in space are more systematic, integrality and scientific.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN912.3

【相似文献】