【摘要】 提出了一套完整的复杂噪声环境下非实时语音信号端点检测系统。首先进行基于样本幅度的预分类处理,然后使用门限过零率算法处理整体幅度很小的样本;整体幅度较大的样本使用引入了自适应噪声估计策略和段落包络幅度特征的改进算法。实际测试表明,在复杂噪声环境下此系统正确率达90.7%,可满足实际系统需求。更多还原

【Abstract】 A voice activity detection system is proposed.There is a pre-processing module to decide which process will be implemented.At the same,two novel features are introduced into the system:adaptive noise estimation and segment amplitude envelope.The VAD system is evaluated by samples obtained from actual exams,and the result shows that the system has a high performance in complicated noise environment.更多还原