期刊文献+

基于独立成分分析技术的语音除噪系统 被引量:2

A Voice Denoising System Based on the Independent Component Analysis
下载PDF
导出
摘要 提出了一种基于独立成分分析技术的语音除噪方法.语音信号与其所夹杂的噪声之间,通常都是相互独立的,采集的观测信号是语音信号与噪声之间的混合信号,通过分析混合语音信号特征,建立了基于独立成分分析技术的语音信号模型.以最大化信号独立性特征为依据,分离语音信号与噪声信号.最后利用多选一信号选择器,实现去噪后语音信号的提取,从而达到语音除噪的目的.对系统的语音去噪效果进行了实例仿真分析,研究表明,该系统可达到理想的去噪效果. A voice denoising method based on independent component analysis was introduced.Usually,the voice is independent of the noise,and the sampled signal is the mixture of the voice and the noise.Based on the independent component analysis technology,a signal processing model was set up by analyzing the characters of the mixture signal.The voice and noise were separated by maximizing the signal's independent characters.At last,a multiplexer was used to extract the voice signal and filtered the noise signal.Example analysis was held to verify the system performance,and the simulation results show that this system has better denoising effect.
出处 《湖南师范大学自然科学学报》 CAS 北大核心 2011年第3期24-26,共3页 Journal of Natural Science of Hunan Normal University
基金 湖南省教育厅科研基金资助项目(10C0922) 湖南师范大学教学教改基金资助项目
关键词 语音除噪 独立成分分析 语音特征 噪声特征 voice denoising independent component analysis voice character noise character
  • 引文网络
  • 相关文献

参考文献8

二级参考文献55

共引文献39

同被引文献21

引证文献2

二级引证文献12

;
使用帮助 返回顶部