基于预处理VAD技术和KL变换的语音增强算法研究

Study of Speech Enhancement Algorithm Based on preprocessing VAD and KL Transform

语音增强的目的是从带噪语音信号中压缩背景噪声 ,提取纯净语音 ,改进通话质量。本文提出了一种基于预处理 VAD技术和自适应 KL变换的语音增强算法。实验表明 ,该算法对于单通道输入有色噪声干扰下的带噪语音信号有较好的增强效果。

The aim of speech enhancement is to compress background noise, to extract pure speech, and to improve communication quality in noisy environment. A new speech enhancement algorithm is proposed, which is named adaptive KL transform speech enhancement algorithm with preprocessing VAD. The simulation results show that the algorithm has better performance for single input channel in environment of colored noise.