A Perceptual Approach to Reduce Musical Noise Phenomenon with Wiener Denoising Technique

Traditional denoising techniques, powerful in terms of noise reduction, have the drawback of generating an annoying musical noise. This paper addresses the problem of enhancing speech in highly noisy environments using perceptual considerations. The postprocessing technique we develop, considers the masking threshold of both noisy speech and the denoised one, to detect musical noise components. Next, to make them inaudible, detected musical noise candidates are set under the noise masking threshold and their closest neighbors are smoothed. Extensive subjective and objective tests have shown that, after enhancement, the musical noise is well reduced even at very low signal to noise ratios