Adaptive wavelet packet thresholding with iterative Kalman filter for speech enhancement
Mengjiao Zhao, Wei‐Ping Zhu · 2017
In this paper, we propose an adaptive wavelet packet (WP) thresholding method with iterative Kalman filter (IKF) for speech enhancement. The WP transform is first applied to the noise corrupted speech on a frame-by-frame basis, which decomposes each frame into a number of subbands. For each subband, a voice activity detector (VAD) is designed to detect the voiced/unvoiced parts of the speech. Based on the VAD result, an adaptive thresholding scheme is then utilized to each subband speech to obtain the pre-enhanced speech. To achieve a further level of enhancement, an IKF is next applied to the pre-enhanced speech. The proposed method is evaluated under various noise conditions. Experimental results are provided to demonstrate the effectiveness of the proposed method as compared to some previous works in terms of segmental SNR and perceptual evaluation of speech quality (PESQ) as two well-known performance indexes.