文档介绍：内蒙古科技大学
本科生毕业设计说明书(毕业论文)
题目:语音增强的应用——基于小波变换的噪声抑制
学生姓名:
学号:2002103235
专业:电子信息工程
班级:电信2003-4班
指导教师:
语音增强的应用——基于小波变换的噪声抑制
摘要
语音信号是非平稳的、时变的、复杂的、信息量大的信号,是语义信息加个人特征的混合信号。语音增强的目的就是要去除语音信号中的噪声和干扰、改善它的质量。
大多数实际应用环境中总存在各种各样的噪声,由于训练环境与识别环境不匹配,现有的绝大多数语音识别系统在噪声环境中的性能都不可避免的急剧下降。
通过小波变换可以有效的抑制周期性噪声、冲激噪声、宽带噪声、语音干扰等各种噪声。对于夹杂在实际语音中的各种噪声能够有很好的抑制效果并明显的增强语音质量。
关键词:语音增强;小波变换;噪声抑制;预处理
The Application of Speech Enhancement– the noise abatement based on the wavelet transformation
Abstract
The speech signal is unbalanced, fluctuant, complex and has a huge amount of information. It is a mixed signal which posed of semantic information and individual characteristic. The aim of the speech enhancement is removing the noise and the disturbance of the speech signal and improving its quality.
The various noises always exist in environment during the practical application. The performance of most existing speech recognition systems usually have a sharp drop inevitably in the noise environment, because the training environment and the recognition environment does not match.
All kinds of noises, such as periodic noise, impulse noise, wide band noise and the speech disturbance, can be effectively suppressed through the wavelet transformation. The wavelet transformation has a good suppression effect on all kinds of noises mixed in the actual speech and could enhance the speech quality significantly.
Key words: Speech enhancement; Wavelet transformation;
Noise abatement; Pretreatment
目录
摘要 I
ABSTRACT II
引言 1
第一章预处理 2
2
2
4
5
6
第二章语音增强概述 7
7
8
8
9
10
11
11
第三章基于小波变换的语音信号增强方法 13
13
14
14
16
18
19
19
22
23
23
总结 27
参考文献 28
附录 29
致谢 32
引言
语言是人类进行思想,观点和情感交流最自然便捷的交互方式,而不同语