THE COMPARATIVE ANALYSIS THE FOURIER TRANSFORM, COSINE TRANSFORM AND WAVELET TRANSFORM AS A SPECTRAL ANALYSIS OF THE DIGITAL SPEECH SIGNALS
DOI:
https://doi.org/10.18372/2310-5461.27.9387Keywords:
wavelet transform, Fourier transform, cosine transform, compression digital speech signals, spectral analysis of the digital speech signals, compression ratio, correlation coefficient, orthogonal wavelet functionsAbstract
The wavelet transform method is proposed to use in the digital speech compression algorithms. The comparative analysis was performed between the Fourier transform, cosine transform and wavelet transform. The feasibility of using wavelet transform unlike Fourier and cosine transform as a spectral analysis of the digital speech signals is grounded and experimentally proved. The evaluating compression ratio was performed depending on the correlation coefficient, the signal-to-noise ratio, peak signal-to-noise ratio and the mean square error. A comparative analysis was performed between the most well-known orthogonal wavelet functions. The experiment result show the superiority of Daubechies and Symlet wavelet functions among all the other studied orthogonal wavelet functions. The results allow to conclude further feasibility of the proposed method of spectral analysis in the digital speech compression algorithms.
References
Daubechies I. Ten Lectures on Wavelets / I. Dau-bechies. — Philadelphia: SIAM, 1992. — 357 s.
Mallat S. A Theory for Multiresolution Signal Decomposition: The Wavelet Representation / S. Mal-lat // IEEE Pattern Anal. and Machine Intell. — 1989. — № 11. — S. 674–693.
Meyer Y. Wavelets and operators / Y. Meyer, 1992. — 244 с. — (Cambridge Studies in Advanced Mathematics).
Shensa M. J. The discrete wavelet transform: wedding the a trous and Mallat Algorithms / M. J. Shensa // IEEE Trans. on Signal Processing. — 1992. — № 40. — S. 2464–2482.
Strang G. Wavelets and Filter Banks / G. Strang, T. Nguyen., 1996. — 490 s.
