Search

We propose a multichannel-to-multichannel target sound extraction (M2M-TSE) framework for separating multichannel target signals from a multichannel mixture of sound sources. Target sound extraction (TSE) isolates a specific target signal using user-provided clues, typically focusing on single-channel extraction with class labels or temporal activation maps. However, to preserve and utilize spatial information in multichannel audio signals, it is essential to extract multichannel signals of a target sound source. Moreover, the clue for extraction can also include spatial or temporal cues like direction-of-arrival (DoA) or timestamps of source activation. To address these challenges, we present an M2M framework that extracts a multichannel sound signal based on spatio-temporal clues. We demonstrate that our transformer-based architecture can successively accomplish the M2M-TSE task for multichannel signals synthesized from audio signals of diverse classes in different room environments. Furthermore, we show that the multichannel extraction task introduces sufficient inductive bias in the DNN, allowing it to directly handle DoA clues without utilizing hand-crafted spatial features., Comment: 5 pages, 4 figures

We present a transformer-based speech-declipping model that effectively recovers clipped signals across a wide range of input signal-to-distortion ratios (SDRs). While recent time-domain deep neural network (DNN)-based declippers have outperformed traditional handcrafted and spectrogram-based DNN approaches, they still struggle with low-SDR inputs. To address this, we incorporate a transformer-based architecture that operates in the time-frequency (TF) domain. The TF-transformer architecture has demonstrated remarkable performance in the speech enhancement task for low-SDR signals but cannot be optimal for the time-domain artifact like clipping. To overcome the limitations of spectrogram-based DNNs, we design an extra convolutional block that directly extracts temporal features from time-domain waveforms. The joint analysis of complex spectrogram and learned temporal features allows the model to improve performance on both high- and low-SDR inputs. Our approach also preserves the unclipped portions of the speech signal during processing, preventing degradation typically seen when only spectral information is used. In evaluations on the VoiceBank-DEMAND and DNS challenge datasets, the proposed model consistently outperformed state-of-the-art (SOTA) declipping models across various metrics, demonstrating its robustness and generalizability., Comment: 5 pages, 2 figures, submitted to ICASSP 2024

This paper presents a framework for universal sound separation and polyphonic audio classification, addressing the challenges of separating and classifying individual sound sources in a multichannel mixture. The proposed framework, DeFT-Mamba, utilizes the dense frequency-time attentive network (DeFTAN) combined with Mamba to extract sound objects, capturing the local time-frequency relations through gated convolution block and the global time-frequency relations through position-wise Hybrid Mamba. DeFT-Mamba surpasses existing separation and classification networks by a large margin, particularly in complex scenarios involving in-class polyphony. Additionally, a classification-based source counting method is introduced to identify the presence of multiple sources, outperforming conventional threshold-based approaches. Separation refinement tuning is also proposed to improve performance further. The proposed framework is trained and tested on a multichannel universal sound separation dataset developed in this work, designed to mimic realistic environments with moving sources and varying onsets and offsets of polyphonic events., Comment: 5 pages, 2 figures

Room geometry inference (RGI) aims at estimating room shapes from measured room impulse responses (RIRs) and has received lots of attention for its importance in environment-aware audio rendering and virtual acoustic representation of a real venue. A lot of estimation models utilizing time difference of arrival (TDoA) or time of arrival (ToA) information in RIRs have been proposed. However, an estimation model should be able to handle more general features and complex relations between reflections to cope with various room shapes and uncertainties such as the unknown number of walls. In this study, we propose a deep neural network that can estimate various room shapes without prior assumptions on the shape or number of walls. The proposed model consists of three sub-networks: a feature extractor, parameter estimation, and evaluation networks, which extract key features from RIRs, estimate parameters, and evaluate the confidence of estimated parameters, respectively. The network is trained by about 40,000 RIRs simulated in rooms of different shapes using a single source and spherical microphone array and tested for rooms of unseen shapes and dimensions. The proposed algorithm achieves almost perfect accuracy in finding the true number of walls and shows negligible errors in room shapes., Comment: 5 pages, 2 figures, Proceedings of the 24th International Congress on Acoustics

Sound event localization and detection (SELD) is a task for the classification of sound events and the localization of direction of arrival (DoA) utilizing multichannel acoustic signals. Prior studies employ spectral and channel information as the embedding for temporal attention. However, this usage limits the deep neural network from extracting meaningful features from the spectral or spatial domains. Therefore, our investigation in this paper presents a novel framework termed the Channel-Spectro-Temporal Transformer (CST-former) that bolsters SELD performance through the independent application of attention mechanisms to distinct domains. The CST-former architecture employs distinct attention mechanisms to independently process channel, spectral, and temporal information. In addition, we propose an unfolded local embedding (ULE) technique for channel attention (CA) to generate informative embedding vectors including local spectral and temporal information. Empirical validation through experimentation on the 2022 and 2023 DCASE Challenge task3 datasets affirms the efficacy of employing attention mechanisms separated across each domain and the benefit of ULE, in enhancing SELD performance., Comment: Accepted to ICASSP 2024

Research finds a weak causal relationship between student loan debt and homeownership. Still, less access to generational wealth among young Black adults is a root cause of higher student debt burden and a substantial barrier to accessing homeownership. The inability to repay debt, or to build wealth through homeownership, in turn, contributes to the persistence and widening of the racial wealth gap. In this report, we describe policy efforts to mitigate the effects of student loan debt on mortgage lending and homeownership and discuss the impact they have on closing the racial homeownership gap. These efforts include flexibility in treatment of student loan debt for mortgage lending, merging student loans into mortgage debt, and state programs to subsidize borrowers' down payments. The current policies will have only marginal impacts on Black homeownership. To continue to lower barriers to homeownership for student loan borrowers and reduce the barriers created by student loan debt overall, we urge consideration of additional interventions. We also encourage policymakers to examine, and possibly pilot, special purpose credit programs aimed at Black student loan borrowers who are otherwise mortgage ready. Finally, we recognize that both the decline in Black homeownership and the burden of student loan debt on Black borrowers are side effects of a broader multigenerational racial wealth gap that has been influenced by both direct and indirect racial discrimination. Ambitious policies aimed at closing this gap overall, outside of the mortgage and student lending spheres, are also needed to reduce disparities within these two areas. [This report was produced by the Urban Institute's Housing Finance Policy Center with funding provided by the Federal Home Loan (FHL) Bank of San Francisco.]

Accurate estimation of indoor space geometries is vital for constructing precise digital twins, whose broad industrial applications include navigation in unfamiliar environments and efficient evacuation planning, particularly in low-light conditions. This study introduces EchoScan, a deep neural network model that utilizes acoustic echoes to perform room geometry inference. Conventional sound-based techniques rely on estimating geometry-related room parameters such as wall position and room size, thereby limiting the diversity of inferable room geometries. Contrarily, EchoScan overcomes this limitation by directly inferring room floorplans and heights, thereby enabling it to handle rooms with arbitrary shapes, including curved walls. The key innovation of EchoScan is its ability to analyze the complex relationship between low- and high-order reflections in room impulse responses (RIRs) using a multi-aggregation module. The analysis of high-order reflections also enables it to infer complex room shapes when echoes are unobservable from the position of an audio device. Herein, EchoScan was trained and evaluated using RIRs synthesized from complex environments, including the Manhattan and Atlanta layouts, employing a practical audio device configuration compatible with commercial, off-the-shelf devices. Compared with vision-based methods, EchoScan demonstrated outstanding geometry estimation performance in rooms with various shapes., Comment: 9 pages, 8 figures, 2 tables

Unsupervised anomalous sound detection (ASD) aims to identify anomalous sounds by learning the features of normal operational sounds and sensing their deviations. Recent approaches have focused on the self-supervised task utilizing the classification of normal data, and advanced models have shown that securing representation space for anomalous data is important through representation learning yielding compact intra-class and well-separated intra-class distributions. However, we show that conventional approaches often fail to ensure sufficient intra-class compactness and exhibit angular disparity between samples and their corresponding centers. In this paper, we propose a training technique aimed at ensuring intra-class compactness and increasing the angle gap between normal and abnormal samples. Furthermore, we present an architecture that extracts features for important temporal regions, enabling the model to learn which time frames should be emphasized or suppressed. Experimental results demonstrate that the proposed method achieves the best performance giving 0.90%, 0.83%, and 2.16% improvement in terms of AUC, pAUC, and mAUC, respectively, compared to the state-of-the-art method on DCASE 2020 Challenge Task2 dataset., Comment: Submitted to ICASSP 2024

Room geometry is important prior information for implementing realistic 3D audio rendering. For this reason, various room geometry inference (RGI) methods have been developed by utilizing the time-of-arrival (TOA) or time-difference-of-arrival (TDOA) information in room impulse responses (RIRs). However, the conventional RGI technique poses several assumptions, such as convex room shapes, the number of walls known in priori, and the visibility of first-order reflections. In this work, we introduce the RGI-Net which can estimate room geometries without the aforementioned assumptions. RGI-Net learns and exploits complex relationships between low-order and high-order reflections in RIRs and, thus, can estimate room shapes even when the shape is non-convex or first-order reflections are missing in the RIRs. RGI-Net includes the evaluation network that separately evaluates the presence probability of walls, so the geometry inference is possible without prior knowledge of the number of walls., Comment: 5 pages, 3 figures, 3 tables

In this work, we present DeFTAN-II, an efficient multichannel speech enhancement model based on transformer architecture and subgroup processing. Despite the success of transformers in speech enhancement, they face challenges in capturing local relations, reducing the high computational complexity, and lowering memory usage. To address these limitations, we introduce subgroup processing in our model, combining subgroups of locally emphasized features with other subgroups containing original features. The subgroup processing is implemented in several blocks of the proposed network. In the proposed split dense blocks extracting spatial features, a pair of subgroups is sequentially concatenated and processed by convolution layers to effectively reduce the computational complexity and memory usage. For the F- and T-transformers extracting temporal and spectral relations, we introduce cross-attention between subgroups to identify relationships between locally emphasized and non-emphasized features. The dual-path feedforward network then aggregates attended features in terms of the gating of local features processed by dilated convolutions. Through extensive comparisons with state-of-the-art multichannel speech enhancement models, we demonstrate that DeFTAN-II with subgroup processing outperforms existing methods at significantly lower computational complexity. Moreover, we evaluate the model's generalization capability on real-world data without fine-tuning, which further demonstrates its effectiveness in practical scenarios., Comment: 13 pages, 6 figures, submitted to IEEE/ACM Trans. Audio, Speech, Lang. Process

Localizing sounds and detecting events in different room environments is a difficult task, mainly due to the wide range of reflections and reverberations. When training neural network models with sounds recorded in only a few room environments, there is a tendency for the models to become overly specialized to those specific environments, resulting in overfitting. To address this overfitting issue, we propose divided spectro-temporal attention. In comparison to the baseline method, which utilizes a convolutional recurrent neural network (CRNN) followed by a temporal multi-head self-attention layer (MHSA), we introduce a separate spectral attention layer that aggregates spectral features prior to the temporal MHSA. To achieve efficient spectral attention, we reduce the frequency pooling size in the convolutional encoder of the baseline to obtain a 3D tensor that incorporates information about frequency, time, and channel. As a result, we can implement spectral attention with channel embeddings, which is not possible in the baseline method dealing with only temporal context in the RNN and MHSA layers. We demonstrate that the proposed divided spectro-temporal attention significantly improves the performance of sound event detection and localization scores for real test data from the STARSS23 development dataset. Additionally, we show that various data augmentations, such as frameshift, time masking, channel swapping, and moderate mix-up, along with the use of external data, contribute to the overall improvement in SELD performance.

The drone has been used for various purposes, including military applications, aerial photography, and pesticide spraying. However, the drone is vulnerable to external disturbances, and malfunction in propellers and motors can easily occur. To improve the safety of drone operations, one should detect the mechanical faults of drones in real-time. This paper proposes a sound-based deep neural network (DNN) fault classifier and drone sound dataset. The dataset was constructed by collecting the operating sounds of drones from microphones mounted on three different drones in an anechoic chamber. The dataset includes various operating conditions of drones, such as flight directions (front, back, right, left, clockwise, counterclockwise) and faults on propellers and motors. The drone sounds were then mixed with noises recorded in five different spots on the university campus, with a signal-to-noise ratio (SNR) varying from 10 dB to 15 dB. Using the acquired dataset, we train a DNN classifier, 1DCNN-ResNet, that classifies the types of mechanical faults and their locations from short-time input waveforms. We employ multitask learning (MTL) and incorporate the direction classification task as an auxiliary task to make the classifier learn more general audio features. The test over unseen data reveals that the proposed multitask model can successfully classify faults in drones and outperforms single-task models even with less training data., Comment: Accepted at 29th International Congress on Sound and Vibration (ICSV29). Dataset available: https://zenodo.org/record/7779574#.ZEVncnZBwQ-

In-situ classification of faulty sounds is an important issue in machine health monitoring and diagnosis. However, in a noisy environment such as a factory, machine sound is always mixed up with environmental noises, and noise-only periods can exist when a machine is not in operation. Therefore, a deep neural network (DNN)-based fault classifier has to be able to distinguish noise from machine sound and be robust to mixed noises. To deal with these problems, we investigate on-site noise exposure (ONE) that exposes a DNN model to the noises recorded in the same environment where the machine operates. Like the outlier exposure technique, noise exposure trains a DNN classifier to produce a uniform predicted probability distribution against noise-only data. During inference, the DNN classifier trained by ONE outputs the maximum softmax probability as the noise score and determines the noise-only period. We mix machine sound and noises of the ToyADMOS2 dataset to simulate highly noisy data. A ResNet-based classifier trained by ONE is evaluated and compared with those trained by other out-of-distribution detection techniques. The test results show that exposing a model to on-site noises can make a model more robust than using other noises or detection techniques., Comment: Presented at 24th International Congress on Acoustics (ICA) 2022

Early degradation prediction of lithium-ion batteries is crucial for ensuring safety and preventing unexpected failure in manufacturing and diagnostic processes. Long-term capacity trajectory predictions can fail due to cumulative errors and noise. To address this issue, this study proposes a data-centric method that uses early single-cycle data to predict the capacity degradation trajectory of lithium-ion cells. The method involves predicting a few knots at specific retention levels using a deep learning-based model and interpolating them to reconstruct the trajectory. Two approaches are used to identify the retention levels of two to four knots: uniformly dividing the retention up to the end of life and finding optimal locations using Bayesian optimization. The proposed model is validated with experimental data from 169 cells using five-fold cross-validation. The results show that mean absolute percentage errors in trajectory prediction are less than 1.60% for all cases of knots. By predicting only the cycle numbers of at least two knots based on early single-cycle charge and discharge data, the model can directly estimate the overall capacity degradation trajectory. Further experiments suggest using three-cycle input data to achieve robust and efficient predictions, even in the presence of noise. The method is then applied to predict various shapes of capacity degradation patterns using additional experimental data from 82 cells. The study demonstrates that collecting only the cycle information of a few knots during model training and a few early cycle data points for predictions is sufficient for predicting capacity degradation. This can help establish appropriate warranties or replacement cycles in battery manufacturing and diagnosis processes.

Deep neural network (DNN) pruning has become a de facto component for deploying on resource-constrained devices since it can reduce memory requirements and computation costs during inference. In particular, channel pruning gained more popularity due to its structured nature and direct savings on general hardware. However, most existing pruning approaches utilize importance measures that are not directly related to the task utility. Moreover, few in the literature focus on visual detection models. To fill these gaps, we propose a novel gradient-based saliency measure for visual detection and use it to guide our channel pruning. Experiments on the KITTI and COCO traffic datasets demonstrate our pruning method's efficacy and superiority over state-of-the-art competing approaches. It can even achieve better performance with fewer parameters than the original model. Our pruning also demonstrates great potential in handling small-scale objects., Comment: 6 pages, 4 figures

In this study, we propose a dense frequency-time attentive network (DeFT-AN) for multichannel speech enhancement. DeFT-AN is a mask estimation network that predicts a complex spectral masking pattern for suppressing the noise and reverberation embedded in the short-time Fourier transform (STFT) of an input signal. The proposed mask estimation network incorporates three different types of blocks for aggregating information in the spatial, spectral, and temporal dimensions. It utilizes a spectral transformer with a modified feed-forward network and a temporal conformer with sequential dilated convolutions. The use of dense blocks and transformers dedicated to the three different characteristics of audio signals enables more comprehensive enhancement in noisy and reverberant environments. The remarkable performance of DeFT-AN over state-of-the-art multichannel models is demonstrated based on two popular noisy and reverberant datasets in terms of various metrics for speech quality and intelligibility., Comment: 5 pages, 2 figures, 3 tables. This article has been published by IEEE Signal Processing Letters. This version is the authors' version and may vary from the final publication in details

Visual detection is a key task in autonomous driving, and it serves as a crucial foundation for self-driving planning and control. Deep neural networks have achieved promising results in various visual tasks, but they are known to be vulnerable to adversarial attacks. A comprehensive understanding of deep visual detectors' vulnerability is required before people can improve their robustness. However, only a few adversarial attack/defense works have focused on object detection, and most of them employed only classification and/or localization losses, ignoring the objectness aspect. In this paper, we identify a serious objectness-related adversarial vulnerability in YOLO detectors and present an effective attack strategy targeting the objectness aspect of visual detection in autonomous vehicles. Furthermore, to address such vulnerability, we propose a new objectness-aware adversarial training approach for visual detection. Experiments show that the proposed attack targeting the objectness aspect is 45.17% and 43.50% more effective than those generated from classification and/or localization losses on the KITTI and COCO traffic datasets, respectively. Also, the proposed adversarial defense approach can improve the detectors' robustness against objectness-oriented attacks by up to 21% and 12% mAP on KITTI and COCO traffic, respectively., Comment: Accepted by 2022 IEEE Intelligent Vehicles Symposium (IV 2022)