Toon van Waterschoot

Randall Ali, Thomas Dietzen, Matteo Scerbo et al.

We introduce a new framework for room acoustics modelling based on a state-space model of the boundary integral equation representing the sound field in a room. Whereas state-space models of linear time-invariant systems are traditionally constructed by means of a state vector and a 4-tuple of system matrices, the state-space representation introduced in this work consists of a state function representing the pressure distribution at the room boundary, and a 4-tuple of integral operators. We refer to this representation as a boundary integral operator state-space (BIOSS) model and provide a physical interpretation for each of the integral operators. As many mathematical operations on vectors and matrices translate to functions and operators, the BIOSS representation can be manipulated to obtain two transfer function representations, having either a feedback or a parallel feedforward structure. Consequently, various equivalent representations for room acoustics are obtained in the BIOSS framework, in the time or frequency domain, and in continuous or discrete space. We discuss two future directions for how the proposed framework can be fertile for research on room acoustics modelling. Firstly, we identify equivalences between the BIOSS framework and various existing room acoustics models (boundary element models, delay networks, geometric models), which may be used to establish relations between existing models and to develop novel room acoustics models. Secondly, we postulate on how concepts from state-space theory, such as observability, controllability, and state realization, can be used for developing new inference and control methods for room acoustics.

Paul Didier, Toon van Waterschoot, Simon Doclo et al.

In a wireless acoustic sensor network (WASN), devices (i.e., nodes) can collaborate through distributed algorithms to collectively perform audio signal processing tasks. This paper focuses on the distributed estimation of node-specific desired speech signals using network-wide Wiener filtering. The objective is to match the performance of a centralized system that would have access to all microphone signals, while reducing the communication bandwidth usage of the algorithm. Existing solutions, such as the distributed adaptive node-specific signal estimation (DANSE) algorithm, converge towards the multichannel Wiener filter (MWF) which solves a centralized linear minimum mean square error (LMMSE) signal estimation problem. However, they do so iteratively, which can be slow and impractical. Many solutions also assume that all nodes observe the same set of sources of interest, which is often not the case in practice. To overcome these limitations, we propose the distributed multichannel Wiener filter (dMWF) for fully connected WASNs. The dMWF is non-iterative and optimal even when nodes observe different sets of sources. In this algorithm, nodes exchange neighbor-pair-specific, low-dimensional (fused) signals estimating the contribution of sources observed by both nodes in the pair. We formally prove the optimality of dMWF and demonstrate its performance in simulated speech enhancement experiments. The proposed algorithm is shown to outperform DANSE in terms of objective metrics after short operation times, highlighting the benefit of its iterationless design.

Paul Didier, Pourya Behmandpoor, Henri Gode et al.

This paper introduces the topology-independent distributed multichannel Wiener filter (TI-dMWF), a novel algorithm for distributed node-specific signal estimation in wireless acoustic sensor networks (WASNs) with unconstrained topologies. The TI-dMWF enables each node in the network to compute its centralized multichannel Wiener filter solution by exchanging only low-dimensional fused signals, without requiring iterative estimation, unlike state-of-the-art approaches such as the topology-independent distributed adaptive node-specific signal estimation (TI-DANSE) algorithm. The TI-dMWF is proven optimal when each source is observed by either all nodes or only one node. Theoretical analysis and numerical simulations confirm that it achieves centralized estimation performance in a single run. Its latency as a function of the pruned-tree depth and its computational complexity are also analyzed. Its robustness is assessed in reverberant-room simulations under estimated second-order statistics, various network topologies, and deviations from the assumed observability model.

Thomas Dietzen, Marc Moonen, Toon van Waterschoot

Power spectral density (PSD) estimates of various microphone signal components are essential to many speech enhancement procedures. As speech is highly non-nonstationary, performance improvements may be gained by maintaining time-variations in PSD estimates. In this paper, we propose an instantaneous PSD estimation approach based on generalized principal components. Similarly to other eigenspace-based PSD estimation approaches, we rely on recursive averaging in order to obtain a microphone signal correlation matrix estimate to be decomposed. However, instead of estimating the PSDs directly from the temporally smooth generalized eigenvalues of this matrix, yielding temporally smooth PSD estimates, we propose to estimate the PSDs from newly defined instantaneous generalized eigenvalues, yielding instantaneous PSD estimates. The instantaneous generalized eigenvalues are defined from the generalized principal components, i.e. a generalized eigenvector-based transform of the microphone signals. We further show that the smooth generalized eigenvalues can be understood as a recursive average of the instantaneous generalized eigenvalues. Simulation results comparing the multi-channel Wiener filter (MWF) with smooth and instantaneous PSD estimates indicate better speech enhancement performance for the latter. A MATLAB implementation is available online.

Stefano Giacomelli, Stefano Damiano, Claudia Rinaldi et al.

This report investigates the extension of pretrained General-Purpose Audio Tagging (GP-AT) models toward spatially grounded Sound Event Localization and Detection (SELD). The proposed AT2SELD framework couples a pretrained AT backbone with compact First-Order Ambisonics (FOA) spatial processing, track-wise SED and Cartesian DOA estimation, permutation aware supervision, and calibration. It characterizes how semantic audio priors support localization-aware scene analysis under data, computation, and deployment constraints. The framework is developed through informed multi-stage Neural Architecture Search (NAS). Stage 1 shows that spectral FOA descriptors, based on magnitude, phase, and Intensity Vectors (IVs), provide the most reliable interface for semantic-to-spatial transfer. Stage 2 identifies early residual spatial encoding as the main capacity-sensitive component, while late track-wise abstraction and recurrent smoothing act mainly as refinement stages. Stage 3 shows that late cross-stitch coupling improves semantic-spatial interaction, whereas early fusion is costlier and less effective. Diagnostic evaluation analyzes the selected architecture under class balancing, focal loss, activity-conditioned DOA supervision, threshold calibration, and transfer across STARSS23, TAU2019, TAU-NIGENS2020, and TAU-NIGENS2021. Focal loss improves the activity point, active-only DOA supervision mitigates inactive target dominance, and validation-selected thresholds recover calibration without replacing spatial learning. Cross-dataset and oracle-activity analyses indicate strong fixed source localization on TAU2019, transferable representations from TAU NIGENS2021, and meaningful but uncertain behavior on STARSS23. Overall, GP-AT priors appear promising for SELD design when embedded in spatial-aware architectures and optimized through integrated calibration and deployment oriented strategies.

Thomas Dietzen, Enzo De Sena, Toon van Waterschoot

The steered response power (SRP) approach to acoustic source localization computes a map of the acoustic scene from the frequency-weighted output power of a beamformer steered towards a set of candidate locations. Equivalently, SRP may be expressed in terms of time-domain generalized cross-correlations (GCCs) at lags equal to the candidate locations' time-differences of arrival (TDOAs). Due to the dense grid of candidate locations, each of which requires inverse Fourier transform (IFT) evaluations, conventional SRP exhibits a high computational complexity. In this paper, we propose a low-complexity SRP approach based on Nyquist-Shannon sampling. Noting that on the one hand the range of possible TDOAs is physically bounded, while on the other hand the GCCs are bandlimited, we critically sample the GCCs around their TDOA interval and approximate the SRP map by interpolation. In usual setups, the number of sample points can be orders of magnitude less than the number of candidate locations and frequency bins, yielding a significant reduction of IFT computations at a limited interpolation cost. Simulations comparing the proposed approximation with conventional SRP indicate low approximation errors and equal localization performance. MATLAB and Python implementations are available online.

T. Dietzen, S. Doclo, M. Moonen et al.

In multi-microphone speech enhancement, reverberation as well as additive noise and/or interfering speech are commonly suppressed by deconvolution and spatial filtering, e.g., using multi-channel linear prediction (MCLP) on the one hand and beamforming, e.g., a generalized sidelobe canceler (GSC), on the other hand. In this paper, we consider several reverberant speech components, whereof some are to be dereverberated and others to be canceled, as well as a diffuse (e.g., babble) noise component to be suppressed. In order to perform both deconvolution and spatial filtering, we integrate MCLP and the GSC into a novel architecture referred to as integrated sidelobe cancellation and linear prediction (ISCLP), where the sidelobe-cancellation (SC) filter and the linear prediction (LP) filter operate in parallel, but on different microphone signal frames. Within ISCLP, we estimate both filters jointly by means of a single Kalman filter. We further propose a spectral Wiener gain post-processor, which is shown to relate to the Kalman filter's posterior state estimate. The presented ISCLP Kalman filter is benchmarked against two state-of-the-art approaches, namely first a pair of alternating Kalman filters respectively performing dereverberation and noise reduction, and second an MCLP+GSC Kalman filter cascade. While the ISCLP Kalman filter is roughly $M^2$ times less expensive than both reference algorithms, where $M$ denotes the number of microphones, it is shown to perform similarly as compared to the former, and to outperform the latter.

T. Dietzen, S. Doclo, M. Moonen et al.

Multi-channel short-time Fourier transform (STFT) domain-based processing of reverberant microphone signals commonly relies on power-spectral-density (PSD) estimates of early source images, where early refers to reflections contained within the same STFT frame. State-of-the-art approaches to multi-source early PSD estimation, given an estimate of the associated relative early transfer functions (RETFs), conventionally minimize the approximation error defined with respect to the early correlation matrix, requiring non-negative inequality constraints on the PSDs. Instead, we here propose to factorize the early correlation matrix and minimize the approximation error defined with respect to the early-correlation-matrix square root. The proposed minimization problem -- constituting a generalization of the so-called orthogonal Procrustes problem -- seeks a unitary matrix and the square roots of the early PSDs up to an arbitrary complex argument, making non-negative inequality constraints redundant. A solution is obtained iteratively, requiring one singular value decomposition (SVD) per iteration. The estimated unitary matrix and early PSD square roots further allow to recursively update the RETF estimate, which is not inherently possible in the conventional approach. An estimate of the said early-correlation-matrix square root itself is obtained by means of the generalized eigenvalue decomposition (GEVD), where we further propose to restore non-stationarities by desmoothing the generalized eigenvalues in order to compensate for inevitable recursive averaging. Simulation results indicate fast convergence of the proposed multi-source early PSD estimation approach in only one iteration if initialized appropriately, and better performance as compared to the conventional approach.

Gert Dekkers, Fernando Rosas, Steven Lauwereins et al.

Smart sensing is expected to become a pervasive technology in smart cities and environments of the near future. These services are improving their capabilities due to integrated devices shrinking in size while maintaining their computational power, which can run diverse Machine Learning algorithms and achieve high performance in various data-processing tasks. One attractive sensor modality to be used for smart sensing are acoustic sensors, which can convey highly informative data while keeping a moderate energy consumption. Unfortunately, the energy budget of current wireless sensor networks is usually not enough to support the requirements of standard microphones. Therefore, energy efficiency needs to be increased at all layers --- sensing, signal processing and communication --- in order to bring wireless smart acoustic sensors into the market. To help to attain this goal, this paper introduces WASN-EM: an energy consumption model for wireless acoustic sensors networks (WASN), whose aim is to aid in the development of novel techniques to increase the energy-efficient of smart wireless acoustic sensors. This model provides a first step of exploration prior to custom design of a smart wireless acoustic sensor, and also can be used to compare the energy consumption of different protocols.

Gert Dekkers, Lode Vuegen, Toon van Waterschoot et al.

The DCASE 2018 Challenge consists of five tasks related to automatic classification and detection of sound events and scenes. This paper presents the setup of Task 5 which includes the description of the task, dataset and the baseline system. In this task, it is investigated to which extent multi-channel acoustic recordings are beneficial for the purpose of classifying domestic activities. The goal is to exploit spectral and spatial cues independent of sensor location using multi-channel audio. For this purpose we provided a development and evaluation dataset which are derivatives of the SINS database and contain domestic activities recorded by multiple microphone arrays. The baseline system, based on a Neural Network architecture using convolutional and dense layer(s), is intended to lower the hurdle to participate the challenge and to provide a reference performance.

Pablo Peso Parada, Dushyant Sharma, Toon van Waterschoot et al.

We present a single channel data driven method for non-intrusive estimation of full-band reverberation time and full-band direct-to-reverberant ratio. The method extracts a number of features from reverberant speech and builds a model using a recurrent neural network to estimate the reverberant acoustic parameters. We explore three configurations by including different data and also by combining the recurrent neural network estimates using a support vector machine. Our best method to estimate DRR provides a Root Mean Square Deviation (RMSD) of 3.84 dB and a RMSD of 43.19 % for T60 estimation.