摘  要在自然界中，语音很容易被外界噪音所扰乱。这些现实存在的噪音不但会降低语音的辨识度与质量，还会很大程度地降低语音处理系统的工作性能。语音增强处理技术不仅能够用来抑制噪声,增大信号的信噪比和提高系统的抗噪声能力，还可以改良语音的音质、可辨度。语音增强还可以作为一个预处理或者是前端处理模块放置在语音处理系统里面。72718

    本论文先从语音的特征、人耳的感知能力、噪声的特点还有语音信号的分析方法着手，对语音信号的短时分析技术进行了简单的分析介绍。本文主要探究了基于谱减法、增强后的谱减法与wiener滤波法的语音增强，从理论方面介绍了谱减法与wiener滤波法，通过公式举例与推算说明各个算法的特点。论文还通过MATLAB对各个算法结果进行了仿真。此外，论文还提及了基于DSP的语音增强处理系统，简单介绍了型号为的DSP芯片构成的语音处理系统，并基于谱减法实现语音的增强。本论文侧重于语音增强算法的仿真，所以对基于DSP的语音增强系统只做了简单的介绍。

毕业论文关键词： 语音增强  谱减法  wiener滤波法  TMS320VC5416

Speech Enhancement Technology

Abstract In nature, the speech can easily be disturbed by outside noise。 The reality of the noise will not only reduce the degree of recognition and voice quality, but also to a large extent reduce the operating performance of the speech processing system。 Speech enhancement processing technology not only can be used to suppress noise, increased noise immunity and improve the signal to noise ratio of the system, it can also improve voice quality, discernible degree。 Speech enhancement can also be used as a pretreatment or a front-end processing module in the Voice Processing System inside。

In this paper, starting with voice features, analysis of the human ear's perception, as well as the noise characteristics of the speech signal to proceed on short-term analysis of the speech signal analysis for a simple introduction。 This paper explores based on spectral subtraction, speech spectral subtraction and wiener filtering method enhanced enhancement, from theory describes the spectral subtraction and wiener filtering method, by way of example and calculate the formula described characteristics of each algorithm。 Thesis for each algorithm through MATLAB simulation results。 In addition, the paper also referred to the DSP-based speech enhancement processing system, a brief introduction to the Voice Processing System Model DSP chip configuration, based on spectral subtraction and voice enhancement。 This paper focuses on the speech enhancement algorithm simulation, so the DSP-based voice enhancement system only a simple introduction。

Key words: Speech Enhancement  Spectral subtraction  Wiener filtering TMS320VC5416

目   录

摘  要 I

Abstract II

目   录 III

图清单 V

变量注释表 VI

1 绪论 1

1。1 课题研究现状与主要内容 1

1。2 语音增强的历史和发展现状 2

1。3 课题研究意义 2

1。4 论文结构 3

2 语音增强的基本知识 4

2。1 语音的产生 4

2。2 人的听觉特性 4