Perceptual Properties Based Signal Subspace Microphone Array Speech Enhancement Algorithm
Perceptual Properties Based Signal Subspace Microphone Array Speech Enhancement Algorithm
复制标题
DOI：
10.3724/sp.j.1004.2009.01481
复制
发表时间：
2010-03
期刊：
影响因子：
--
通讯作者：
Ning Cheng;Wenju Liu
中科院分区：
文献类型：
--
作者：
Ning Cheng;Wenju Liu
针对麦克风阵列信号子空间语音增强算法的不足, 结合人耳的听觉掩蔽效应, 提出了改进的信号子空间算法.利用置信度函数确定噪声子空间的维数。在噪声子空间中，利用条件似然法估计噪声功率谱。然后，将掩蔽特性引入子空间方法中，对线性滤波器进行估计。实验结果表明，与传统算法相比，该算法能更有效地抑制噪声，语音质量客观指标得到显著提高。
To aim at the drawbacks of the conventional subspace microphone array speech enhancement method, some improvements are proposed by using masking properties of human ears. Confldence function is used to determine the noise subspace dimension. In noise subspace, the noise power spectrum is estimated by the conditional likelihood. Then, the masking properties are incorporated into the subspace method to estimate the linear fllter. Experiments show that compared with conventional algorithms, the proposed approach suppresses noise more efiectively and obtains a signiflcant improvement on objective speech quality measures.