宋倩倩,于凤芹.基于Hilbert-Huang变换和听觉掩蔽的语音增强算法[J].声学技术,2009,(3):280~283 |
基于Hilbert-Huang变换和听觉掩蔽的语音增强算法 |
Speech enhancement based on Hilbert-Huang tansform and human auditory masking |
投稿时间:2008-07-17 修订日期:2008-10-21 |
DOI: |
中文关键词: Hilbert-Huang变换 经验模态分解 固有模态函数 听觉掩蔽 语音增强 |
英文关键词: HHT empirical mode decomposition intrinsic mode function human auditory masking speech enhancement |
基金项目: |
|
摘要点击次数: 1327 |
全文下载次数: 1732 |
中文摘要: |
Hilbert-Huang变换是一种新型的具有自适应性的时频分析方法,分析了HHT算法的原理,提出了一种基于HHT和听觉掩蔽的语音增强算法,首先将语音信号进行EMD分解得到各阶IMF分量,然后对高频IMF分量进行听觉掩蔽处理,最后将处理后的分量与剩余分量叠加得到重构信号。仿真结果表明所提出的算法降低了语音失真测度值,提高了语音信号的信噪比、清晰度及可懂度,并与听觉掩蔽算法和谱减法进行了比较,显示了该算法的优越性。 |
英文摘要: |
HHT is a new and self-adaptable method for time-frequency analysis.The theory of HHT is studied and a speech enhancement method based on HHT and human auditory masking is brought forward.First the signal is decomposed into IMFs with the method of EMD,then the high-frequency IMFs is processed with the human auditory masking,finally the signal is reconstructed by adding the treated IMFs with the residual IMFs.Simulation experiments show that it can reduce the measured value of speech distortion and improve the SNR,speech articulation and intelligibility.The results display the superiority of this method over the human auditory masking and the spectral subtraction method. |
查看全文
查看/发表评论 下载PDF阅读器 |
关闭 |
|
|
|