Stochastic approach for noise suppression of speech signal by considering finite range of amplitude fluctuation in real environment

Abstract

Several noise suppression methods for speech signal have been proposed up to now. On the other hand, the actual speech signal fluctuates within a finite range and the observed data sometimes are affected by amplitude saturation owing to the existence of definite dynamic range in measurement instrument. In this study, a signal processing method to suppress the noise for actual speech signal is proposed in an appropriate form for the finite amplitude range of the measured data. More specifically, a new type of noise suppression method is proposed by introducing a statistical orthogonal expansion expression of the probability distribution based on beta distribution defined within the finite fluctuation ranges of speech signal and observation. Furthermore, the effectiveness of the proposed method is confirmed experimentally by applying it to the actual speech signal contaminated by noise.