arXiv subjects

Sharon Gannot

Publications and source records attributed to Sharon Gannot.

At least 55 records · Page 3Linked to original sources

arXiv2021

Semi-supervised source localization in reverberant environments with deep generative modeling

We propose a semi-supervised approach to acoustic source localization in reverberant environments based on deep generative modeling. Localization in reverberant environments remains an open challenge. Even with large data volumes, the number of labels available for supervised learning in reverberant environments is usually small. We address this issue by performing semi-supervised learning (SSL) with convolutional variational autoencoders (VAEs) on reverberant speech signals recorded with microphone arrays. The VAE is trained to generate the phase of relative transfer functions (RTFs) between microphones, in parallel with a direction of arrival (DOA) classifier based on RTF-phase. These models are trained using both labeled and unlabeled RTF-phase sequences. In learning to perform these tasks, the VAE-SSL explicitly learns to separate the physical causes of the RTF-phase (i.e., source location) from distracting signal characteristics such as noise and speech activity. Relative to existing semi-supervised localization methods in acoustics, VAE-SSL is effectively an end-to-end processing approach which relies on minimal preprocessing of RTF-phase features. As far as we are aware, our paper presents the first approach to modeling the physics of acoustic propagation using deep generative modeling. The VAE-SSL approach is compared with two signal processing-based approaches, steered response power with phase transform (SRP-PHAT) and MUltiple SIgnal Classification (MUSIC), as well as fully supervised CNNs. We find that VAE-SSL can outperform the conventional approaches and the CNN in label-limited scenarios. Further, the trained VAE-SSL system can generate new RTF-phase samples, which shows the VAE-SSL approach learns the physics of the acoustic environment. The generative modeling in VAE-SSL thus provides a means of interpreting the learned representations.

eess.SP↗

arXiv2021

Semi-supervised source localization with deep generative modeling

We propose a semi-supervised localization approach based on deep generative modeling with variational autoencoders (VAEs). Localization in reverberant environments remains a challenge, which machine learning (ML) has shown promise in addressing. Even with large data volumes, the number of labels available for supervised learning in reverberant environments is usually small. We address this issue by performing semi-supervised learning (SSL) with convolutional VAEs. The VAE is trained to generate the phase of relative transfer functions (RTFs), in parallel with a DOA classifier, on both labeled and unlabeled RTF samples. The VAE-SSL approach is compared with SRP-PHAT and fully-supervised CNNs. We find that VAE-SSL can outperform both SRP-PHAT and CNN in label-limited scenarios.

eess.AS↗

arXiv2021

Speech enhancement with mixture-of-deep-experts with clean clustering pre-training

In this study we present a mixture of deep experts (MoDE) neural-network architecture for single microphone speech enhancement. Our architecture comprises a set of deep neural networks (DNNs), each of which is an 'expert' in a different speech spectral pattern such as phoneme. A gating DNN is responsible for the latent variables which are the weights assigned to each expert's output given a speech segment. The experts estimate a mask from the noisy input and the final mask is then obtained as a weighted average of the experts' estimates, with the weights determined by the gating DNN. A soft spectral attenuation, based on the estimated mask, is then applied to enhance the noisy speech signal. As a byproduct, we gain reduction at the complexity in test time. We show that the experts specialization allows better robustness to unfamiliar noise types.

cs.SD↗

arXiv2020

Multichannel Online Dereverberation based on Spectral Magnitude Inverse Filtering

This paper addresses the problem of multichannel online dereverberation. The proposed method is carried out in the short-time Fourier transform (STFT) domain, and for each frequency band independently. In the STFT domain, the time-domain room impulse response is approximately represented by the convolutive transfer function (CTF). The multichannel CTFs are adaptively identified based on the cross-relation method, and using the recursive least square criterion. Instead of the complex-valued CTF convolution model, we use a nonnegative convolution model between the STFT magnitude of the source signal and the CTF magnitude, which is just a coarse approximation of the former model, but is shown to be more robust against the CTF perturbations. Based on this nonnegative model, we propose an online STFT magnitude inverse filtering method. The inverse filters of the CTF magnitude are formulated based on the multiple-input/output inverse theorem (MINT), and adaptively estimated based on the gradient descent criterion. Finally, the inverse filtering is applied to the STFT magnitude of the microphone signals, obtaining an estimate of the STFT magnitude of the source signal. Experiments regarding both speech enhancement and automatic speech recognition are conducted, which demonstrate that the proposed method can effectively suppress reverberation, even for the difficult case of a moving speaker.

cs.SD↗

arXiv2020

Misalignment Recognition in Acoustic Sensor Networks using a Semi-supervised Source Estimation Method and Markov Random Fields

In this paper, we consider the problem of acoustic source localization by acoustic sensor networks (ASNs) using a promising, learning-based technique that adapts to the acoustic environment. In particular, we look at the scenario when a node in the ASN is displaced from its position during training. As the mismatch between the ASN used for learning the localization model and the one after a node displacement leads to erroneous position estimates, a displacement has to be detected and the displaced nodes need to be identified. We propose a method that considers the disparity in position estimates made by leave-one-node-out (LONO) sub-networks and uses a Markov random field (MRF) framework to infer the probability of each LONO position estimate being aligned, misaligned or unreliable while accounting for the noise inherent to the estimator. This probabilistic approach is advantageous over naive detection methods, as it outputs a normalized value that encapsulates conditional information provided by each LONO sub-network on whether the reading is in misalignment with the overall network. Experimental results confirm that the performance of the proposed method is consistent in identifying compromised nodes in various acoustic conditions.

eess.AS↗

arXiv2020

FCN Approach for Dynamically Locating Multiple Speakers

In this paper, we present a deep neural network-based online multi-speaker localisation algorithm. Following the W-disjoint orthogonality principle in the spectral domain, each time-frequency (TF) bin is dominated by a single speaker, and hence by a single direction of arrival (DOA). A fully convolutional network is trained with instantaneous spatial features to estimate the DOA for each TF bin. The high resolution classification enables the network to accurately and simultaneously localize and track multiple speakers, both static and dynamic. Elaborated experimental study using both simulated and real-life recordings in static and dynamic scenarios, confirms that the proposed algorithm outperforms both classic and recent deep-learning-based algorithms.

eess.AS↗

arXiv2020

Binaural LCMV Beamforming with Partial Noise Estimation

Besides reducing undesired sources (interfering sources and background noise), another important objective of a binaural beamforming algorithm is to preserve the spatial impression of the acoustic scene, which can be achieved by preserving the binaural cues of all sound sources. While the binaural minimum variance distortionless response (BMVDR) beamformer provides a good noise reduction performance and preserves the binaural cues of the desired source, it does not allow to control the reduction of the interfering sources and distorts the binaural cues of the interfering sources and the background noise. Hence, several extensions have been proposed. First, the binaural linearly constrained minimum variance (BLCMV) beamformer uses additional constraints, enabling to control the reduction of the interfering sources while preserving their binaural cues. Second, the BMVDR with partial noise estimation (BMVDR-N) mixes the output signals of the BMVDR with the noisy reference microphone signals, enabling to control the binaural cues of the background noise. Merging the advantages of both extensions, in this paper we propose the BLCMV with partial noise estimation (BLCMV-N). We show that the output signals of the BLCMV-N can be interpreted as a mixture of the noisy reference microphone signals and the output signals of a BLCMV using an adjusted interference scaling parameter. We provide a theoretical comparison between the BMVDR, the BLCMV, the BMVDR-N and the proposed BLCMV-N in terms of noise and interference reduction performance and binaural cue preservation. Experimental results using recorded signals as well as the results of a perceptual listening test show that the BLCMV-N is able to preserve the binaural cues of an interfering source (like the BLCMV), while enabling to trade off between noise reduction performance and binaural cue preservation of the background noise (like the BMVDR-N).

eess.AS↗

arXiv2020

ML Estimation and CRBs for Reverberation, Speech and Noise PSDs in Rank-Deficient Noise-Field

Speech communication systems are prone to performance degradation in reverberant and noisy acoustic environments. Dereverberation and noise reduction algorithms typically require several model parameters, e.g. the speech, reverberation and noise power spectral densities (PSDs). A commonly used assumption is that the noise PSD matrix is known. However, in practical acoustic scenarios, the noise PSD matrix is unknown and should be estimated along with the speech and reverberation PSDs. In this paper, we consider the case of rank-deficient noise PSD matrix, which arises when the noise signal consists of multiple directional interference sources, whose number is less than the number of microphones. We derive two closed-form maximum likelihood estimators (MLEs). The first is a non-blocking-based estimator which jointly estimates the speech, reverberation and noise PSDs, and the second is a blocking-based estimator, which first blocks the speech signal and then jointly estimates the reverberation and noise PSDs. Both estimators are analytically compared and analyzed, and mean square errors (MSEs) expressions are derived. Furthermore, Cramer-Rao Bounds (CRBs) on the estimated PSDs are derived. The proposed estimators are examined using both simulation and real reverberant and noisy signals, demonstrating the advantage of the proposed method compared to competing estimators.

eess.AS↗

arXiv2019

Machine learning in acoustics: theory and applications

Acoustic data provide scientific and engineering insights in fields ranging from biology and communications to ocean and Earth science. We survey the recent advances and transformative potential of machine learning (ML), including deep learning, in the field of acoustics. ML is a broad family of techniques, which are often based in statistics, for automatically detecting and utilizing patterns in data. Relative to conventional acoustics and signal processing, ML is data-driven. Given sufficient training data, ML can discover complex relationships between features and desired labels or actions, or between features themselves. With large volumes of training data, ML can discover models describing complex acoustic phenomena such as human speech and reverberation. ML in acoustics is rapidly developing with compelling results and significant future promise. We first introduce ML, then highlight ML developments in four acoustics research areas: source localization in speech processing, source localization in ocean acoustics, bioacoustics, and environmental sounds in everyday scenes.

eess.SP↗

arXiv2019

MIRaGe: Multichannel Database Of Room Impulse Responses Measured On High-Resolution Cube-Shaped Grid In Multiple Acoustic Conditions

We introduce a database of multi-channel recordings performed in an acoustic lab with adjustable reverberation time. The recordings provide information about room impulse responses (RIR) for various positions of a loudspeaker. In particular, the main positions correspond to 4104 vertices of a cube-shaped dense grid within a 46x36x32 cm volume. The database thus provides a tool for detailed analyses of beampatterns of spatial processing methods as well as for training and testing of mathematical models of the acoustic field.

eess.AS↗

arXiv2019

Speech Dereverberation Using Fully Convolutional Networks

Speech derverberation using a single microphone is addressed in this paper. Motivated by the recent success of the fully convolutional networks (FCN) in many image processing applications, we investigate their applicability to enhance the speech signal represented by short-time Fourier transform (STFT) images. We present two variations: a "U-Net" which is an encoder-decoder network with skip connections and a generative adversarial network (GAN) with U-Net as generator, which yields a more intuitive cost function for training. To evaluate our method we used the data from the REVERB challenge, and compared our results to other methods under the same conditions. We have found that our method outperforms the competing methods in most cases.

eess.AS↗

arXiv2019

Deep Clustering Based on a Mixture of Autoencoders

In this paper we propose a Deep Autoencoder MIxture Clustering (DAMIC) algorithm based on a mixture of deep autoencoders where each cluster is represented by an autoencoder. A clustering network transforms the data into another space and then selects one of the clusters. Next, the autoencoder associated with this cluster is used to reconstruct the data-point. The clustering algorithm jointly learns the nonlinear data representation and the set of autoencoders. The optimal clustering is found by minimizing the reconstruction loss of the mixture of autoencoder network. Unlike other deep clustering algorithms, no regularization term is needed to avoid data collapsing to a single point. Our experimental evaluations on image and text corpora show significant improvement over state-of-the-art methods.

cs.LG↗

arXiv2018

Data-Driven Source Separation Based on Simplex Analysis

Blind source separation (BSS) is addressed, using a novel data-driven approach, based on a well-established probabilistic model. The proposed method is specifically designed for separation of multichannel audio mixtures. The algorithm relies on spectral decomposition of the correlation matrix between different time frames. The probabilistic model implies that the column space of the correlation matrix is spanned by the probabilities of the various speakers across time. The number of speakers is recovered by the eigenvalue decay, and the eigenvectors form a simplex of the speakers' probabilities. Time frames dominated by each of the speakers are identified exploiting convex geometry tools on the recovered simplex. The mixing acoustic channels are estimated utilizing the identified sets of frames, and a linear umixing is performed to extract the individual speakers. The derived simplexes are visually demonstrated for mixtures of 2, 3 and 4 speakers. We also conduct a comprehensive experimental study, showing high separation capabilities in various reverberation conditions.

eess.AS↗

arXiv2018

Multichannel Speech Separation and Enhancement Using the Convolutive Transfer Function

This paper addresses the problem of speech separation and enhancement from multichannel convolutive and noisy mixtures, \emph{assuming known mixing filters}. We propose to perform the speech separation and enhancement task in the short-time Fourier transform domain, using the convolutive transfer function (CTF) approximation. Compared to time-domain filters, CTF has much less taps, consequently it has less near-common zeros among channels and less computational complexity. The work proposes three speech-source recovery methods, namely: i) the multichannel inverse filtering method, i.e. the multiple input/output inverse theorem (MINT), is exploited in the CTF domain, and for the multi-source case, ii) a beamforming-like multichannel inverse filtering method applying single source MINT and using power minimization, which is suitable whenever the source CTFs are not all known, and iii) a constrained Lasso method, where the sources are recovered by minimizing the $\ell_1$-norm to impose their spectral sparsity, with the constraint that the $\ell_2$-norm fitting cost, between the microphone signals and the mixing model involving the unknown source signals, is less than a tolerance. The noise can be reduced by setting a tolerance onto the noise power. Experiments under various acoustic conditions are carried out to evaluate the three proposed methods. The comparison between them as well as with the baseline methods is presented.

cs.SD↗

arXiv2017

Multiple-Speaker Localization Based on Direct-Path Features and Likelihood Maximization with Spatial Sparsity Regularization

This paper addresses the problem of multiple-speaker localization in noisy and reverberant environments, using binaural recordings of an acoustic scene. A Gaussian mixture model (GMM) is adopted, whose components correspond to all the possible candidate source locations defined on a grid. After optimizing the GMM-based objective function, given an observed set of binaural features, both the number of sources and their locations are estimated by selecting the GMM components with the largest priors. This is achieved by enforcing a sparse solution, thus favoring a small number of speakers with respect to the large number of initial candidate source locations. An entropy-based penalty term is added to the likelihood, thus imposing sparsity over the set of GMM priors. In addition, the direct-path relative transfer function (DP-RTF) is used to build robust binaural features. The DP-RTF, recently proposed for single-source localization, was shown to be robust to reverberations, since it encodes inter-channel information corresponding to the direct-path of sound propagation. In this paper, we extend the DP-RTF estimation to the case of multiple sources. In the short-time Fourier transform domain, a consistency test is proposed to check whether a set of consecutive frames is associated to the same source or not. Reliable DP-RTF features are selected from the frames that pass the consistency test to be used for source localization. Experiments carried out using both simulation data and real data gathered with a robotic head confirm the efficiency of the proposed multi-source localization method.

cs.SD↗

arXiv2017

Speech Enhancement using a Deep Mixture of Experts

In this study we present a Deep Mixture of Experts (DMoE) neural-network architecture for single microphone speech enhancement. By contrast to most speech enhancement algorithms that overlook the speech variability mainly caused by phoneme structure, our framework comprises a set of deep neural networks (DNNs), each one of which is an 'expert' in enhancing a given speech type corresponding to a phoneme. A gating DNN determines which expert is assigned to a given speech segment. A speech presence probability (SPP) is then obtained as a weighted average of the expert SPP decisions, with the weights determined by the gating DNN. A soft spectral attenuation, based on the SPP, is then applied to enhance the noisy speech signal. The experts and the gating components of the DMoE network are trained jointly. As part of the training, speech clustering into different subsets is performed in an unsupervised manner. Therefore, unlike previous methods, a phoneme-labeled database is not required for the training procedure. A series of experiments with different noise types verified the applicability of the new algorithm to the task of speech enhancement. The proposed scheme outperforms other schemes that either do not consider phoneme structure or use a simpler training methodology.

cs.SD↗

arXiv2017

Joint Alignment of Multiple Point Sets with Batch and Incremental Expectation-Maximization

This paper addresses the problem of registering multiple point sets. Solutions to this problem are often approximated by repeatedly solving for pairwise registration, which results in an uneven treatment of the sets forming a pair: a model set and a data set. The main drawback of this strategy is that the model set may contain noise and outliers, which negatively affects the estimation of the registration parameters. In contrast, the proposed formulation treats all the point sets on an equal footing. Indeed, all the points are drawn from a central Gaussian mixture, hence the registration is cast into a clustering problem. We formally derive batch and incremental EM algorithms that robustly estimate both the GMM parameters and the rotations and translations that optimally align the sets. Moreover, the mixture's means play the role of the registered set of points while the variances provide rich information about the contribution of each component to the alignment. We thoroughly test the proposed algorithms on simulated data and on challenging real data collected with range sensors. We compare them with several state-of-the-art algorithms, and we show their potential for surface reconstruction from depth data.

cs.CV↗

arXiv2016

Semi-Supervised Source Localization on Multiple-Manifolds with Distributed Microphones

The problem of source localization with ad hoc microphone networks in noisy and reverberant enclosures, given a training set of prerecorded measurements, is addressed in this paper. The training set is assumed to consist of a limited number of labelled measurements, attached with corresponding positions, and a larger amount of unlabelled measurements from unknown locations. However, microphone calibration is not required. We use a Bayesian inference approach for estimating a function that maps measurement-based feature vectors to the corresponding positions. The central issue is how to combine the information provided by the different microphones in a unified statistical framework. To address this challenge, we model this function using a Gaussian process with a covariance function that encapsulates both the connections between pairs of microphones and the relations among the samples in the training set. The parameters of the process are estimated by optimizing a maximum likelihood (ML) criterion. In addition, a recursive adaptation mechanism is derived where the new streaming measurements are used to update the model. Performance is demonstrated for 2-D localization of both simulated data and real-life recordings in a variety of reverberation and noise levels.

cs.SD↗