Binaural Heterophasic Superdirective Beamforming

The superdirective beamformer, while attractive for processing broadband acoustic signals, often suffers from the problem of white noise amplification. So, its application requires well-designed acoustic arrays with sensors of extremely low self-noise level, which is difficult if not impossible to attain. In this paper, a new binaural superdirective beamformer is proposed, which is divided into two sub-beamformers. Based on studies and facts in psychoacoustics, these two filters are designed in such a way that they are orthogonal to each other to make the white noise components in the binaural beamforming outputs incoherent while maximizing the output interaural coherence of the diffuse noise, which is important for the brain to localize the sound source of interest. As a result, the signal of interest in the binaural superdirective beamformer’s outputs is in phase but the white noise components in the outputs are random phase, so the human auditory system can better separate the acoustic signal of interest from white noise by listening to the outputs of the proposed approach. Experimental results show that the derived binaural superdirective beamformer is superior to its conventional monaural counterpart.

Generalized Sidelobe Canceller Beamforming with Combined Postfilter and Sparse NMF for Speech Enhancement

A multi-microphone array speech enhancement method using Generalized Sidelobe Canceller (GSC) beamforming with Combined Postfilter (CP) and Sparse Non-negative Matrix Factorization (SNMF) is proposed in this paper. GSC beamforming with CP and SNMF is implemented to reduce directional noise, diffuse noise, residual noise and to separate interferences in adverse environment. In this paper, the directional noise is reduced using GSC beamforming, whereas the diffuse noise in each subband is reduced with a combined postfilter using Unconstrained Frequency domain Normalized Least Mean Square (UFNLMS) algorithm. Finally, the residual noise at the output of CP is eliminated by SNMF which optimizes the noise. The performance of the proposed method is evaluated using parameters like PESQ, SSNR, STOI, SDR and LSD. The noise reduction for four and eight microphones is compared and illustrated in spectrograms. The proposed method shows better performance in terms of intelligibility and quality when compared to the existing methods in adverse environments.

A Coherence Based Dual-Microphone Speech Enhancement in Diffuse Noise Field

In recent years, the problem of improving the quality of noisy signal has always been a dilemma, challenging scientists; especially in diffuse noise conditions. There has been great and growing interest in dual-microphone processing for hands-free speech enhancement. In this paper, the author addresses a improvement of coherence based filter in diffuse noise field. Explicit information about speech presence or absence is necessary in many speech processing applications. The proposal algorithm uses a soft speech presence probability as an optimal parameter to process noisy signal in diffuse noise condition. The evaluation proves that, speech application dual-microphone can be incorporated with this efficiency algorithm for pre-processing

A Speech Enhancement MVDR Algorithm in Diffuse Noise Field

The minimum variance distortionless response (MVDR) beamformer often used in speech application for separating sound source and suppressing ambient, coherent, stationary and non-stationary noise in real complex environment. This paper deals problem speech enhancement in diffuse noise field by using a modified MVDR, which incorporates speech presence probability to estimate auto and cross power spectral densities. This combination gives the advantage of saving target speaker while suppressing background noise. A efficiency post-filtering, which is a function depends on speech presence probability, used for increasing the quality of filtered signal. The performance evaluation demonstrates the ability of proposal algorithm when compared to conventional MVDR.

Perceptual Evaluation of Binaural MVDR-Based Algorithms to Preserve the Interaural Coherence of Diffuse Noise Fields

Besides improving speech intelligibility in background noise, another important objective of noise reduction algorithms for binaural hearing devices is preserving the spatial impression for the listener. In this study, we evaluate the performance of several recently proposed noise reduction algorithms based on the binaural minimum-variance-distortionless-response (MVDR) beamformer, which trade-off between noise reduction performance and preservation of the interaural coherence (IC) for diffuse noise fields. Aiming at a perceptually optimized result, this trade-off is determined based on the IC discrimination ability of the human auditory system. The algorithms are evaluated with normal-hearing participants for an anechoic scenario and a reverberant cafeteria scenario, in terms of both speech intelligibility using a matrix sentence test and spatial quality using a MUlti Stimulus test with Hidden Reference and Anchor (MUSHRA). The results show that all the binaural noise reduction algorithms are able to improve speech intelligibility compared with the unprocessed microphone signals, where partially preserving the IC of the diffuse noise field leads to a significant improvement in perceived spatial quality compared with the binaural MVDR beamformer while hardly affecting speech intelligibility.