A new analysis of perceptual speech enhancement is
presented. It focuses on the fact that if only noise above the masking
threshold is filtered, then noise below the masking threshold, but
above the absolute threshold of hearing, can become audible after the
masker filtering. This particular drawback of some perceptual filters,
hereafter called the maskee-to-audible-noise (MAN) phenomenon,
favours the emergence of isolated tonals that increase musical noise.
Two filtering techniques that avoid or correct the MAN phenomenon
are proposed to effectively suppress background noise without introducing
much distortion. Experimental results, including objective
and subjective measurements, show that these techniques improve
the enhanced speech quality and the gain they bring emphasizes the
importance of the MAN phenomenon.
 N. Virag, "Single channel speech enhancement based on masking properties
of the human auditory system," IEEE Trans. Speech and Audio
Processing, vol. 7, pp. 126-137, 1999.
 Y. Hu and P. Loizou, "Incorporating a psychoacoustic model in frequency
domain speech enhancement," IEEE Signal Processing Letters, vol. 11(2),
pp. 270-273, Feb 2004.
 L. Lin, W. H. Holmes, and E. Ambikairajah, "Speech denoising using
perceptual modification of wiener filtering," IEE Electronic Letters,
vol. 38, pp. 1486-1487, Nov 2002.
 Y. Ephraim and D. Malah, "Speech enhancement using a minimum
mean square error short-time spectral amplitude estimator," IEEE Trans.
Acoust., Speech, Signal Processing, vol. ASSP-32, pp. 1109-1121, Dec
 A. Amehraye, D. Pastor, and A. Tamtaoui, "Perceptual improvement of
wiener filtering." Proc. of ICASSP, 2008, pp. 2081-2084.
 C. Beaugeant, V. Turbin, P. Scalart, and A. Gilloire, "New optimal
filtering approaches for hands-free telecommunication terminals," Signal
Processing, vol. 64, pp. 33-47(15), Jan 1998.
 J. D. Johnston, "Transform coding of audio signals using perceptual noise
criteria," IEEE Jour. Selected Areas Commun, vol. 6, pp. 314-323, 1988.
 ITU-T(2003), "Subjective test methodology for evaluating speech communication
systems that include noise suppression algorithm," ITU-T
Recommendation P.835, 2003.