语音增强的多频带处理技术日益受到重视,提出G_W联合算法,使用Gammatone滤波器组将含噪语音信号分解成若干个频带信号,通过改进的Wiener滤波技术对各个频带信号进行降噪,最后综合这若干频带,得到增强后的语音。实验结果表明,该方法能够更为有效地抑制宽带噪声,降低音乐噪声残留,获得较好的听觉舒适度,具有实用价值。
Multi-band technology for speech enhancement attracts great attention in recent days.A G_W combined algorithm for speech enhancement is proposed in this paper.Speech signals are divided into several bands by the Gamrnatone filter. Each band can be denoised by Wiener filter,and the enhanced speech is obtained by synthesizing all the bands.The effective result shows the good performance of this algorithm,while wide band noise and musical residua are suppressed.