This paper presents a new approach to speech

enhancement based on the H1 filtering. This approach differs

from the traditional modified Wiener/Kalman filtering approach

in the following two aspects: 1) no a priori knowledge of the noise

source statistics is required, the only assumption made is that

noise signals have a finite energy; 2) the estimation criterion for

the filter design is to minimize the worst possible amplification

of the estimation error signals in terms of the modeling errors

and additive noises. Since most additive noises in speech are non-

Gaussian, this estimation approach is highly robust and more

appropriate in practical speech enhancement. The proposed approach

is straightforward to implement, as detailed in this paper.

Experimental results show consistently superior enhancement

performance of the H1 filtering algorithm over the Kalman

filtering counterpart, measured by the global signal-to-noise ratio

(SNR). Examination of the spectrogram displays for the enhanced

speech shows that the H1 filtering approach tends to be more

effective where the assumptions on the noise statistics are less

valid.

PDF file

In: IEEE Trans. on Speech and Audio Processing

Details