Parameterized MMSE spectral magnitude estimation for the enhancement of noisy speech

Abstract

The enhancement of short-term spectra of noisy speech can be achieved by statistical estimation of the clean speech spectral components. We present a minimum mean-square error estimator of the clean speech spectral magnitude that uses both a parametric compression function in the estimation error criterion and a parametric prior distribution for the statistical model of the clean speech magnitude. The novel parametric estimator has many known magnitude estimators as a special solution and, additionally, affords estimators that combine the beneficial properties of different known solutions. The new estimator is evaluated in terms of segmental SNR, speech distortion, and noise suppression.

Speech dereverberation using backward estimation of the late reverberant spectral variance

“…While there are many fidelity criteria it was recently found that the MMSE of the root spectral amplitude provides a good tradeoff between speech distortion, musical noise and noise reduction [8]. The corresponding fidelity criteria is given by…”

Front-end technologies for robust ASR in reverberant environments—spectral enhancement-based dereverberation and auditory modulation filterbank features

“…The constant parameters μ and γ can be tuned to yield several types of estimators. In [21], μ = 0.5 and γ = 0.5 have been identified as a good compromise between the amount of musical noise and the clarity of speech and are, therefore, also applied here. For obtaining the correct approximation for the selected values of μ and γ , the exponents p 0 and p ∞ in (16) have to be set to 0.5 and 1.0, respectively [42].…”

“…Here, the MMSE estimator described in [21] is used to obtain the clean speech magnitude, which requires the PSDs of the speech signal and the interference. It has been shown, e.g., in [3,18], that attenuating late reverberation is crucial for ASR while early reflections can be mitigated well by, e.g., cepstral mean subtraction (CMS) [35].…”

“…In the final step, a parameterized MMSE spectral magnitude estimator [21] is used to determine the weighting function G[ ] to obtain the enhanced speech signalŶ d [ ]. A simplified, computationally less complex version [42] based on the confluent hypergeometric function [43] is used, which is defined as…”

“…Therefore, a clean speech estimator is required which keeps the speech distortions at a low level and introduces a minimal amount of artifacts like musical noise. The minimum mean square error (MMSE) estimator of the clean speech amplitudes proposed in [21] is a parametric estimator that can be tuned in such a way that it gives a good compromise between musical noise, speech distortions, and speech enhancement. For this, reliable estimates of the corresponding desired speech and interference power spectral densities (PSDs) are required.…”