A Proposed Speech Scrambling based on Haar Transform and Permutation

In this paper, the original speech will be scrambling using a hybrid structure of Wavelet- Haar transform and Permutation. The Haar transforms were applied on the speech file after dividing it into frames with 256-samples size. In the first case, the sound frame was passed through a low pass filter and high pass filter at the first level of the analysis, after that in the second level, the low pass filter coefficients input again in a series of low and high pass. In the second case, the sound frame was passed through a low pass filter and high pass filter in the first level of the analysis. The second level, the coefficients of high pass filter passed through a series of low and high pass filters. While for the third case, each frame was converted to a two-dimensional matrix; a 2D-haar transform was applied to the matrix and then returned to the frame with one-dimension. All three cases applying the permutation process at the end of the Wavelet -Haar transform, to offer higher possible security. And all three cases simulated in system with noise channel of multi (SNR) values, and then the security and quality were measured by (SSNR).