Speech signal reconstruction based on the symmetric tight wavelet frame decomposition

This paper presents an approach for speech signal reconstruction using symmetric tight wavelet frame decomposition. The motivation of this approach is the desire of using the advantages of tight wavelet frame into speech signal reconstruction. Compared to wavelet bases the tight wavelet frame is nearly shift-invariant and provide better time-frequency localization, robust reconstruction, wavelet smoothness, short support, and symmetry. To evaluate the performance of reconstructed signal we use the signal to noise ratio SNR, the peak signal to noise ratio PSNR and the normalized root mean square error NRMSE. The experimental results indicate that using tight wavelet frame significantly improves the quality of reconstructed signal.