Cyclic Multichannel Wiener Filter for Acoustic Beamforming

TL;DR

This paper introduces a cyclic multichannel Wiener filter for speech enhancement that leverages the cyclostationary nature of voiced speech, improving MSE performance but sensitive to fundamental frequency estimation.

Contribution

It proposes a novel cyclic multichannel Wiener filter based on cyclostationary speech models, extending traditional methods to better exploit harmonic spectral correlations.

Findings

Significant SI-SDR improvements on synthetic data

High sensitivity to fundamental frequency estimation accuracy

Reduces to traditional MWF for stationary signals

Abstract

Acoustic beamforming models typically assume wide-sense stationarity of speech signals within short time frames. However, voiced speech is better modeled as a cyclostationary (CS) process, a random process whose mean and autocorrelation are $T_1$-periodic, where $\alpha_1=1/T_1$ corresponds to the fundamental frequency of vowels. Higher harmonic frequencies are found at integer multiples of the fundamental. This work introduces a cyclic multichannel Wiener filter (cMWF) for speech enhancement derived from a cyclostationary model. This beamformer exploits spectral correlation across the harmonic frequencies of the signal to further reduce the mean-squared error (MSE) between the target and the processed input. The proposed cMWF is optimal in the MSE sense and reduces to the MWF when the target is wide-sense stationary. Experiments on simulated data demonstrate considerable improvements in scale-invariant signal-to-distortion ratio (SI-SDR) on synthetic data but also indicate high sensitivity to the accuracy of the estimated fundamental frequency $\alpha_1$, which limits effectiveness on real data.