摘要
提出了一种基于EEMD域统计模型的话音激活检测算法。算法首先利用总体平均经验模态分解(Ensemble empirical mode decomposition,EEMD)对带噪语音进行分解,得到信号的本征模式函数(Intrinsicmode function,IMF)分量,选择与原信号的相关性最高的两个分量相加组成主分量;然后对主分量进行频域分解,引入统计模型,求出EEMD域特征参数;最后利用噪声与语音的EEMD域特征参数的不同来进行语音激活检测。实验结果表明,在不同信噪比情况下,本文算法性能优于目前常用的VAD算法,特别在噪声强度大时体现出明显的优势。
Voice activity detection algorithm based on ensemble empirical mode decomposition domain statistical model is presented. The noisy speech is decomposed into intrinsic mode func- tion (IMF) components by using ensemble empirical mode decomposition (EEMD) method. Two IMF components with the higher correlation with original speech are added to calculate the characteristic parameter of the statistical model. The decision of the speech/noise is made by comparing the characteristic parameter with its threshold. The proposed VAD algorithm is tested on speech signals under various noise conditions with several SNRs. Experimental resuits show that the proposed VAD algorithm outperforms some standard VAD algorithms, especially under a low SNR noisy condition.
出处
《数据采集与处理》
CSCD
北大核心
2012年第1期51-56,共6页
Journal of Data Acquisition and Processing
基金
江苏省自然科学基金(BK2009059)资助项目
关键词
话音激活检测
经验模式分解
总体平均经验模式分解
EEMD域统计模型
voice activity detection (VAD)
empirical mode decomposition (EMD)
ensembleempirical mode decomposition (EEMD)
EEMD domain statistical model