Enhanced voice activity detection in kernel subspace domain
Enhanced voice activity detection in kernel subspace domain
复制标题
DOI：
10.1121/1.4809770
复制
发表时间：
2013-07-01
影响因子：
2.4
通讯作者：
Chang, Joon-Hyuk
中科院分区：
文献类型：
--
作者：
Kim, Dong Kook;Shin, Jong Won;Chang, Joon-Hyuk
为了提高基于核的语音活动检测的性能，提出了一种基于核子空间域的语音活动检测方法。提出了一种利用核主成分分析同时对角化两个协方差矩阵的线性变换矩阵来生成核子空间。对核子空间中的VAD进行了基于高斯分布的似然比检验。实验结果表明，该算法在各种噪声条件下均优于传统方法。(C) 2013年美国声学学会
This paper proposes a voice activity detection (VAD) method in a kernel subspace domain to improve the performance of the kernel-based VAD. A linear transform matrix that can simultaneously diagonalize the two covariance matrices using kernel principal component analysis is presented to generate the kernel subspace. The likelihood ratio test based on Gaussian distributions is applied for the VAD in the kernel subspace. Experimental results show that the proposed VAD algorithm outperforms the conventional approaches under various noise conditions. (C) 2013 Acoustical Society of America