大学毕业设计_基于倒谱的大学生语音识别算法研究.docVIP

LANZHOU UNIVERSITY OF TECHNOLOGY 毕业论文 题目： 基于倒谱的大学生语音识别算法研究 College Students?Speech Recognition Algorithm based on?Cepstrum 摘要 语音是人类最重要的交流工具，随着电子计算机和人工智能机器的广泛应用，人们发现人和机器之间最好的通信方式是语言通信，而语音是语言的声学表现形式。要使机器听的懂人话，就要对语音信号进行处理。随着信息技术的不断发展，尤其是网络技术的日益普及和完善，语音信号处理技术正发挥着越来越重要的作用。为了寻找良好性能的特征、提取重要参数以提高识别系统性能，各种科学的算法应运而生。 语音识别的关键技术是提取出语音特征，语音特征有很多，倒谱分析就是其中之一。倒谱法作为信号处理的重要的方法，能够得到比较好的识别性能。本文主要介绍了语音识别技术。基本实现思想是将输入的随机语音信号通过线性变换系统处理为加性信号，用基于Mel频率的倒谱系数（MFCC）以及一些语音信号的固有特征进行倒谱分析。接下来通过对语音倒谱在各个方面的应用进行MATLAB编程仿真，得到语音基音检测和共振峰检测的实验仿真结果。 关键词：倒谱；语音识别；共振峰检测；基音检测；Abstract Voice is one of the most important human communication tools. With the widely application of electronic computers and artificially intellective machine, it was discovered that language communication is the best way to communicate between man and machine and the voice was the reflection of the language. If we wanted us understood by the machines, it is necessary to deal with the signal. With the continuous development of IT, especially the popularity of network and perfect system, voice signal processing technology plays an increasingly important role. In order to find a good performance characteristics and extract important parameters to improve the performance of the recognition system, a variety of scientific algorithms have been emerged. The key technology of voice recognition is extracting voice features. While there are many voice features, cep-strum analyses is one of them. The cep-strum, as one of the important signal processing methods, can get better recognition performance. This paper mainly introduce the voice recognition technology. The basic idea is to casually input voice signal and transform the signal by using additive signal to linear transformation, and use the methods based on Mel(MFCC) frequency as well as some of the speech signal to cep-strum analyze. And then by the usage of the voice spectrum of applications in all aspects programs MATLAB simulation, it can get the