Y. Ghanbari, M.R. Karami-Mollaei (Iran), and B. Amelifard (USA)
Speech processing, speech enhancement, spectralsubtraction
In this paper, we propose a new approach to improve the performance of speech enhancement technique based on multi-band spectral subtraction for white Gaussian noise. First, the original power spectral subtraction and multi band spectral subtraction methods are surveyed and implemented. Next, the generalization is applied on multi band spectral subtraction. Finally, the flattened noise spectrum with the value of the average of spectrum components is used as the noise estimation and then a power factor is optimized in order to increase the enhancement performance. The results show a meaningful improvement. Comparisons of three methods are reported.
Important Links:
Go Back
