一种基于模糊系统的语音增强方法

Research on speech enhancement method based on fuzzy system

语音增强是解决噪声污染的有效方法,它的首要目标是在接收端尽可能从带噪语音中恢复纯净的语音信号.讨论强背景噪声下的基于模糊系统的语音增强方法,并与减谱法语音增强算法进行比较.计算机仿真结果表明,该方法在大大消除背景噪声的同时,消除了音乐噪声,同时保持语音信号较好的可懂度.

Speech enhancement is an efficient method in removing the noise, the aim of which is to retrieval the pure speech signal from the speech signal containing noise. In this paper the speech enhancement method based on fuzzy system is discussed in strong background noise, and a comparison is made between this method and the spectral subtraction algorithm. Experiment results show that this method eliminates the background noise and the music noise simultaneously, while keeps the comprehensibility of the speech signal.