登录

  • 登录
  • 忘记密码?点击找回

注册

  • 获取手机验证码 60
  • 注册

找回密码

  • 获取手机验证码60
  • 找回
毕业论文网 > 毕业论文 > 电子信息类 > 电子信息工程 > 正文

基于先验信噪比估计的维纳滤波语音降噪方法的研究

 2023-02-03 10:02  

论文总字数:17984字

摘 要

在现实生活中,声音传播过程中不可避免的会受到外界的干扰,而语音降噪就是用来处理这一问题的技术。它的最终目的就是将受噪声干扰甚至淹没的语音信号中尽可能的还原出干净的声音信号。而噪音又来自生活中,在进行建筑施工时会产生噪声,在工厂也会有机器的噪声,而随着人民生活水平的提高汽车的普让汽车也产生了大量噪声。所以,人们生活中的噪音处处存在,它严重影响着现代生活中的通话质量。然而又不可能彻底消除它,但至少去想办法还原出有用的语音信号。

因此对于带噪语音信号进行降噪已经成为语音通信系统中十分迫切的问题。由于数字信号相关方面的发展,1970年代掀起一个理论研究浪潮并且取得了一定的进步,以及高速DSP芯片的出现,使语音增强方法开始得以实用化的。

本文基于维纳滤波语音增强技术。首先概要掌握经典语音增强方法,推导单通道维纳滤波公式,然后由分析维纳滤波的原理并实现算法,分析研究从先验信噪比估计,并基于此研究维纳滤波的语音减少噪声的办法。知道语音信号分析中语音增强性能优劣的常用标准,研究了维纳滤波降噪方法的详细过程并且对这个方法进行了改进,并对仿真结果做了简要分析。与原方法的对比实验验证了改进的方法能够有效的提高语音可懂度。

关键词:语音增强;人耳感知;维纳滤波;先验信噪比

Research on Wiener Filter Speech Denoising Method Based on Prior Signal-to-Noise Ratio Estimation

Abstract

In real life, voice transmission process will inevitably be subject to external interference, and voice noise reduction is used to deal with this problem. Its ultimate goal is to find as clean a sound signal as possible in the speech signal disturbed or even submerged by noise. Noise comes from people"s daily life. Noise will be generated during construction and in factories. With the improvement of people"s living standards, the popularity of automobiles also produces noise. Therefore, noise exists everywhere in our life, which seriously affects the quality of our telephone. But we can not completely eliminate it, but at least to find ways to restore useful voice signals.

Therefore, denoising of noisy speech signals has become a very urgent problem in speech communication system. Due to the development of digital signal related aspects, a wave of theoretical research was launched in the 1970s, and some progress was made. With the emergence of high-speed DSP chips, speech enhancement methods began to be practical.

This paper is based on Wiener filtering speech enhancement technology. Firstly, the classical speech enhancement methods are briefly mastered, and the formula of single channel Wiener filtering is deduced. Then, the principle and algorithm of Wiener filtering are analyzed, and the estimation of prior signal-to-noise ratio is analyzed. Based on this, the method of reducing noise by Wiener filtering is studied. Knowing the common standards of speech enhancement performance in speech signal analysis, the detailed process of Wiener filter denoising method is studied, and the method is improved, and the simulation results are briefly analyzed. Compared with the original method, the experimental results show that the improved method can effectively improve the speech intelligibility.

Keywords: speech enhancement;human ear perception;Wiener Filtering;priori SNR

目录

摘要

Abstract

第一章绪论

1.1语音增强研究背景和意义

1.2语音增强算法及其发展现状[4]

1.3.维纳滤波法研究意义

1.4论文主要内容

第二章语音增强的基本原理

2.1语音增强技术概述

2.2语音增强的基础知识

2.2.1语音特性

2.2.2人耳感知特性

2.2.3噪声特性

2.3语音系统的性能评价

2.3.1主观评价方法

2.3.2客观评价方法

第三章基于先验信噪比估计的维纳滤波语音降噪方法的研究

3.1算法原理

3.2改进的维纳滤波法

3.1实验与仿真

第四章全文总结

致谢

参考文献

附录

第一章绪论

1.1语音增强研究背景和意义

现今人们生活在一个有声的世界里,有音乐,有广播,有对话。但往往这些有用的信号又包含在各种各样的噪声中。这些都会让人与人之间的交流和获得需要的声音变得十分困难,甚至会听不清对方在说些什么。从而可能会丢失一些重要的信息造成严重的损失。可见如何嘈杂的环境中获取准确而清晰的声音是多么的重要。所以人们致力于运用科技技术,研究语音增强,将噪声尽可能的降低使生活更美好。

剩余内容已隐藏,请支付后下载全文,论文总字数:17984字

您需要先支付 80元 才能查看全部内容!立即支付
已经是最后一篇了

企业微信

Copyright © 2010-2022 毕业论文网 站点地图