Speech Enhancement using Perceptual Frequency-Weighting

S. Hayashi and M. Suguimoto (Japan)


Speech enhancement, Perceptual weighting, Masking property, Noise reduction, Spectral subtraction


The present paper describes a quality enhancement of speech corrupted by additive background noise in a sin gle channel system. The proposed approach is based on the introduction of a perceptual criteria using a frequency weighting filter in a subtractive-type enhancement process. Although the subtractive-type method is very attractive be cause of its simplicity, it produces an unnatural and un pleasant residual noise. Thus, choosing fixed optimized pa rameters for all speech and noise conditions is difficult. A new and effective algorithm is thus developed based on the masking properties of the human ear. This newly devel oped algorithm allows for an automatic adaptation in the time and frequency of the enhancement system, and finds a suitable noise estimate according to the frequency of the corrupted speech. Experimental results show that the pro posed approach can efficiently remove additive noise re lated to various types of noise corruption.

Important Links:

Go Back