Search

This paper investigates the feasibility of class-incremental learning (CIL) for Sound Event Localization and Detection (SELD) tasks. The method features an incremental learner that can learn new sound classes independently while preserving knowledge of old classes. The continual learning is achieved through a mean square error-based distillation loss to minimize output discrepancies between subsequent learners. The experiments are conducted on the TAU-NIGENS Spatial Sound Events 2021 dataset, which includes 12 different sound classes and demonstrate the efficacy of proposed method. We begin by learning 8 classes and introduce the 4 new classes at next stage. After the incremental phase, the system is evaluated on the full set of learned classes. Results show that, for this realistic dataset, our proposed method successfully maintains baseline performance across all metrics.

Recent advancements in music source separation have significantly progressed, particularly in isolating vocals, drums, and bass elements from mixed tracks. These developments owe much to the creation and use of large-scale, multitrack datasets dedicated to these specific components. However, the challenge of extracting similarly sounding sources from orchestra recordings has not been extensively explored, largely due to a scarcity of comprehensive and clean (i.e bleed-free) multitrack datasets. In this paper, we introduce a novel multitrack dataset called SynthSOD, developed using a set of simulation techniques to create a realistic (i.e. using high-quality soundfonts), musically motivated, and heterogeneous training set comprising different dynamics, natural tempo changes, styles, and conditions. Moreover, we demonstrate the application of a widely used baseline music separation model trained on our synthesized dataset w.r.t to the well-known EnsembleSet, and evaluate its performance under both synthetic and real-world conditions., Comment: Submitted to the OJSP - ICASSP 2025

In this paper, we provide a novel Model-free approach based on Deep Neural Network (DNN) to accomplish point prediction and prediction interval under a general regression setting. Usually, people rely on parametric or non-parametric models to bridge dependent and independent variables (Y and X). However, this classical method relies heavily on the correct model specification. Even for the non-parametric approach, some additive form is often assumed. A newly proposed Model-free prediction principle sheds light on a prediction procedure without any model assumption. Previous work regarding this principle has shown better performance than other standard alternatives. Recently, DNN, one of the machine learning methods, has received increasing attention due to its great performance in practice. Guided by the Model-free prediction idea, we attempt to apply a fully connected forward DNN to map X and some appropriate reference random variable Z to Y. The targeted DNN is trained by minimizing a specially designed loss function so that the randomness of Y conditional on X is outsourced to Z through the trained DNN. Our method is more stable and accurate compared to other DNN-based counterparts, especially for optimal point predictions. With a specific prediction procedure, our prediction interval can capture the estimation variability so that it can render a better coverage rate for finite sample cases. The superior performance of our method is verified by simulation and empirical studies.

Acoustical signal processing of directional representations of sound fields, including source, receiver, and scatterer transfer functions, are often expressed and modeled in the spherical harmonic domain (SHD). Certain such modeling operations, or applications of those models, involve multiplications of those directional quantities, which can also be expressed conveniently in the SHD through coupling coefficients known as Gaunt coefficients. Since the definition and notation of Gaunt coefficients varies across acoustical publications, this work defines them based on established conventions of complex and real spherical harmonics (SHs) along with a convenient matrix form for spherical multiplication of directionally band-limited spherical functions. Additionally, the report provides a derivation of the Gaunt coefficients for real SHs, which has been missing from the literature and can be used directly in spatial audio frameworks such as Ambisonics. Matlab code is provided that can compute all coefficients up to user specified SH orders. Finally, a number of relevant acoustical processing examples from the literature are presented, following the matrix formalism of coefficients introduced in the report.

In end-to-end multi-channel speech enhancement, the traditional approach of designating one microphone signal as the reference for processing may not always yield optimal results. The limitation is particularly in scenarios with large distributed microphone arrays with varying speaker-to-microphone distances or compact, highly directional microphone arrays where speaker or microphone positions change over time. Current mask-based methods often fix the reference channel during training, which makes it not possible to adaptively select the reference channel for optimal performance. To address this problem, we introduce an adaptive approach for selecting the optimal reference channel. Our method leverages a multi-channel masking-based scheme, where multiple masked signals are combined to generate a single-channel output signal. This enhanced signal is then used for loss calculation, while the reference clean speech is adjusted based on the highest scale-invariant signal-to-distortion ratio (SI-SDR). The experimental results on the Spear challenge simulated dataset D4 demonstrate the superiority of our proposed method over the conventional approach of using a fixed reference channel with single-channel masking., Comment: Accepted by EUSIPCO 2024

Deep neural networks (DNN) has received increasing attention in machine learning applications in the last several years. Recently, a non-asymptotic error bound has been developed to measure the performance of the fully connected DNN estimator with ReLU activation functions for estimating regression models. The paper at hand gives a small improvement on the current error bound based on the latest results on the approximation ability of DNN. More importantly, however, a non-random subsampling technique--scalable subsampling--is applied to construct a `subagged' DNN estimator. Under regularity conditions, it is shown that the subagged DNN estimator is computationally efficient without sacrificing accuracy for either estimation or prediction tasks. Beyond point estimation/prediction, we propose different approaches to build confidence and prediction intervals based on the subagged DNN estimator. In addition to being asymptotically valid, the proposed confidence/prediction intervals appear to work well in finite samples. All in all, the scalable subsampling DNN estimator offers the complete package in terms of statistical inference, i.e., (a) computational efficiency; (b) point estimation/prediction accuracy; and (c) allowing for the construction of practically useful confidence and prediction intervals.

Distance estimation from audio plays a crucial role in various applications, such as acoustic scene analysis, sound source localization, and room modeling. Most studies predominantly center on employing a classification approach, where distances are discretized into distinct categories, enabling smoother model training and achieving higher accuracy but imposing restrictions on the precision of the obtained sound source position. Towards this direction, in this paper we propose a novel approach for continuous distance estimation from audio signals using a convolutional recurrent neural network with an attention module. The attention mechanism enables the model to focus on relevant temporal and spectral features, enhancing its ability to capture fine-grained distance-related information. To evaluate the effectiveness of our proposed method, we conduct extensive experiments using audio recordings in controlled environments with three levels of realism (synthetic room impulse response, measured response with convolved speech, and real recordings) on four datasets (our synthetic dataset, QMULTIMIT, VoiceHome-2, and STARSS23). Experimental results show that the model achieves an absolute error of 0.11 meters in a noiseless synthetic scenario. Moreover, the results showed an absolute error of about 1.30 meters in the hybrid scenario. The algorithm's performance in the real scenario, where unpredictable environmental factors and noise are prevalent, yields an absolute error of approximately 0.50 meters. For reproducible research purposes we make model, code, and synthetic datasets available at https://github.com/michaelneri/audio-distance-estimation., Comment: Accepted for publication in IEEE/ACM Transactions on Audio, Speech, and Language Processing

Sound Event Detection and Localization (SELD) is a combined task of identifying sound events and their corresponding direction-of-arrival (DOA). While this task has numerous applications and has been extensively researched in recent years, it fails to provide full information about the sound source position. In this paper, we overcome this problem by extending the task to Sound Event Detection, Localization with Distance Estimation (3D SELD). We study two ways of integrating distance estimation within the SELD core - a multi-task approach, in which the problem is tackled by a separate model output, and a single-task approach obtained by extending the multi-ACCDOA method to include distance information. We investigate both methods for the Ambisonic and binaural versions of STARSS23: Sony-TAU Realistic Spatial Soundscapes 2023. Moreover, our study involves experiments on the loss function related to the distance estimation part. Our results show that it is possible to perform 3D SELD without any degradation of performance in sound event detection and DOA estimation., Comment: This paper has been accepted for the 32nd European Signal Processing Conference EUSIPCO 2024 in Lyon

Scene-based spatial audio formats, such as Ambisonics, are playback system agnostic and may therefore be favoured for delivering immersive audio experiences to a wide range of (potentially unknown) devices. The number of channels required to deliver high spatial resolution Ambisonic audio, however, can be prohibitive for low-bandwidth applications. Therefore, this paper proposes a compression codec, which is based upon the parametric higher-order Directional Audio Coding (HO-DirAC) model. The encoder downmixes the higher-order Ambisonic (HOA) input audio into a reduced number of signals, which are accompanied by perceptually-motivated scene parameters. The downmixed audio is coded using a perceptual audio coder, whereas the parameters are grouped into perceptual bands, quantized, and downsampled. On the decoder side, low Ambisonic orders are fully recovered. Not fully recoverable HOA components are synthesized according to the parameters. The results of a listening test indicate that the proposed parametric spatial audio codec can improve the adopted perceptual audio coder, especially at low to medium-high bitrates, when applied to fifth-order HOA signals., Comment: Accepted for publication in Proceedings of the 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 2024)

Ambisonics encoding of microphone array signals can enable various spatial audio applications, such as virtual reality or telepresence, but it is typically designed for uniformly-spaced spherical microphone arrays. This paper proposes a method for Ambisonics encoding that uses a deep neural network (DNN) to estimate a signal transform from microphone inputs to Ambisonics signals. The approach uses a DNN consisting of a U-Net structure with a learnable preprocessing as well as a loss function consisting of mean average error, spatial correlation, and energy preservation components. The method is validated on two microphone arrays with regular and irregular shapes having four microphones, on simulated reverberant scenes with multiple sources. The results of the validation show that the proposed method can meet or exceed the performance of a conventional signal-independent Ambisonics encoder on a number of error metrics., Comment: Accepted for publication in Proceedings of the 2024 IEEE International Conference on Acoustics, Speech and Signal Processing

Current multichannel speech enhancement algorithms typically assume a stationary sound source, a common mismatch with reality that limits their performance in real-world scenarios. This paper focuses on attention-driven spatial filtering techniques designed for dynamic settings. Specifically, we study the application of linear and nonlinear attention-based methods for estimating time-varying spatial covariance matrices used to design the filters. We also investigate the direct estimation of spatial filters by attention-based methods without explicitly estimating spatial statistics. The clean speech clips from WSJ0 are employed for simulating speech signals of moving speakers in a reverberant environment. The experimental dataset is built by mixing the simulated speech signals with multichannel real noise from CHiME-3. Evaluation results show that the attention-driven approaches are robust and consistently outperform conventional spatial filtering approaches in both static and dynamic sound environments.

To address the difficult problem of multi-step ahead prediction of non-parametric autoregressions, we consider a forward bootstrap approach. Employing a local constant estimator, we can analyze a general type of non-parametric time series model, and show that the proposed point predictions are consistent with the true optimal predictor. We construct a quantile prediction interval that is asymptotically valid. Moreover, using a debiasing technique, we can asymptotically approximate the distribution of multi-step ahead non-parametric estimation by bootstrap. As a result, we can build bootstrap prediction intervals that are pertinent, i.e., can capture the model estimation variability, thus improving upon the standard quantile prediction intervals. Simulation studies are given to illustrate the performance of our point predictions and pertinent prediction intervals for finite samples.

This paper proposes neural networks for compensating sensorineural hearing loss. The aim of the hearing loss compensation task is to transform a speech signal to increase speech intelligibility after further processing by a person with a hearing impairment, which is modeled by a hearing loss model. We propose an interpretable model called dynamic processing network, which has a structure similar to band-wise dynamic compressor. The network is differentiable, and therefore allows to learn its parameters to maximize speech intelligibility. More generic models based on convolutional layers were tested as well. The performance of the tested architectures was assessed using spectro-temporal objective index (STOI) with hearing-threshold noise and hearing aid speech intelligibility (HASPI) metrics. The dynamic processing network gave a significant improvement of STOI and HASPI in comparison to popular compressive gain prescription rule Camfit. A large enough convolutional network could outperform the interpretable model with the cost of larger computational load. Finally, a combination of the dynamic processing network with convolutional neural network gave the best results in terms of STOI and HASPI.

While direction of arrival (DOA) of sound events is generally estimated from multichannel audio data recorded in a microphone array, sound events usually derive from visually perceptible source objects, e.g., sounds of footsteps come from the feet of a walker. This paper proposes an audio-visual sound event localization and detection (SELD) task, which uses multichannel audio and video information to estimate the temporal activation and DOA of target sound events. Audio-visual SELD systems can detect and localize sound events using signals from a microphone array and audio-visual correspondence. We also introduce an audio-visual dataset, Sony-TAu Realistic Spatial Soundscapes 2023 (STARSS23), which consists of multichannel audio data recorded with a microphone array, video data, and spatiotemporal annotation of sound events. Sound scenes in STARSS23 are recorded with instructions, which guide recording participants to ensure adequate activity and occurrences of sound events. STARSS23 also serves human-annotated temporal activation labels and human-confirmed DOA labels, which are based on tracking results of a motion capture system. Our benchmark results demonstrate the benefits of using visual object positions in audio-visual SELD tasks. The data is available at https://zenodo.org/record/7880637., Comment: 27 pages, 9 figures, accepted for publication in NeurIPS 2023 Track on Datasets and Benchmarks

Many multi-source localization and tracking models based on neural networks use one or several recurrent layers at their final stages to track the movement of the sources. Conventional recurrent neural networks (RNNs), such as the long short-term memories (LSTMs) or the gated recurrent units (GRUs), take a vector as their input and use another vector to store their state. However, this approach results in the information from all the sources being contained in a single ordered vector, which is not optimal for permutation-invariant problems such as multi-source tracking. In this paper, we present a new recurrent architecture that uses unordered sets to represent both its input and its state and that is invariant to the permutations of the input set and equivariant to the permutations of the state set. Hence, the information of every sound source is represented in an individual embedding and the new estimates are assigned to the tracked trajectories regardless of their order., Comment: Accepted for publication at Forum Acusticum 2023

The rapid adoption of smart grids demands robust security and efficiency measures due to their critical role in delivering electricity and their potential for customer-oriented benefits. This paper presents an innovative framework, named RETINA, which provides a resilient and secure energy trading mechanism within smart grid systems. RETINA tackles the inherent security and infrastructure challenges in smart grids by establishing a trust-based security layer and facilitating energy transactions through blockchain technology. Our proposed solution integrates Public Key Infrastructure (PKI) and the Web of Trust (WoT) concepts, promoting decentralized communication channels and robust key management. We further introduce a smart contract-based energy trading mechanism that factors in trust, distance, and energy type (green or non-green) in cost calculation. The utility and robustness of RETINA have been validated in a virtualized testbed environment with 500 nodes, demonstrating superior performance in terms of scalability and resilience compared to the existing WoT scheme. Furthermore, RETINA successfully enables a secure and efficient energy trading scheme, promoting the use of renewable energy sources. Future enhancements will include application to a realistic smart grid deployment and the integration of additional functionalities. This groundbreaking solution has the potential to revolutionize the smart grid ecosystem, addressing its current limitations and propelling the industry towards a future of advanced and secure energy exchange., Comment: Under submission to Elsevier's Sustainable Energy, Grids and Networks (SEGAN)

The non-linear autoregressive (NLAR) model plays an important role in modeling and predicting time series. One-step ahead prediction is straightforward using the NLAR model, but the multi-step ahead prediction is cumbersome. For instance, iterating the one-step ahead predictor is a convenient strategy for linear autoregressive (LAR) models, but it is suboptimal under NLAR. In this paper, we first propose a simulation and/or bootstrap algorithm to construct optimal point predictors under an $L_1$ or $L_2$ loss criterion. In addition, we construct bootstrap prediction intervals in the multi-step ahead prediction problem; in particular, we develop an asymptotically valid quantile prediction interval as well as a pertinent prediction interval for future values. In order to correct the undercoverage of prediction intervals with finite samples, we further employ predictive -- as opposed to fitted -- residuals in the bootstrap process. Simulation studies are also given to substantiate the finite sample performance of our methods.

The increase in luminosity, and consequent higher backgrounds, of the LHC upgrades require improved rejection of fake tracks in the forward region of the ATLAS Muon Spectrometer. The New Small Wheel upgrade of the Muon Spectrometer aims to reduce the large background of fake triggers from track segments that are not originated from the interaction point. The New Small Wheel employs two detector technologies, the resistive strip Micromegas detectors and the "small" Thin Gap Chambers, with a total of 2.45 Million electrodes to be sensed. The two technologies require the design of a complex electronics system given that it consists of two different detector technologies and is required to provide both precision readout and a fast trigger. It will operate in a high background radiation region up to about 20 kHz/cm$^{2}$ at the expected HL-LHC luminosity of $\mathcal{L}$=7.5$\times10^{34}$cm$^{-2}$s$^{-1}$. The architecture of the system is strongly defined by the GBTx data aggregation ASIC, the newly-introduced FELIX data router and the software based data handler of the ATLAS detector. The electronics complex of this new detector was designed and developed in the last ten years and consists of multiple radiation tolerant Application Specific Integrated Circuits, multiple front-end boards, dense boards with FPGA's and purpose-built Trigger Processor boards within the ATCA standard. The New Small Wheel has been installed in 2021 and is undergoing integration within ATLAS for LHC Run 3. It should operate through the end of Run 4 (December 2032). In this manuscript, the overall design of the New Small Wheel electronics is presented., Comment: 61 pages

This work proposes a learnable filterbank based on a multi-channel masking framework for multi-channel source separation. The learnable filterbank is a 1D Conv layer, which transforms the raw waveform into a 2D representation. In contrast to the conventional single-channel masking method, we estimate a mask for each individual microphone channel. The estimated masks are then applied to the transformed waveform representation like in the traditional filter-and-sum beamforming operation. Specifically, each mask is used to multiply the corresponding channel's 2D representation, and the masked output of all channels are then summed. At last, a 1D transposed Conv layer is used to convert the summed masked signal into the waveform domain. The experimental results show our method outperforms single-channel masking with a learnable filterbank and can outperform multi-channel complex masking with STFT complex spectrum in the STGCSEN model if a learnable filterbank is transformed to a higher feature dimension. The spatial response analysis also verifies that multi-channel masking in the learnable filterbank domain has spatial selectivity.