本文提出了一个针对Detection and Classification Acoustic Scene and Events 2021 Challenge (DCASE) Task 4提交方案的基准测试，这些方案代表了Sound Event Detection任务中最先进技术的采样。提交方案根据为DCASE 2021 Challenge Task 4提出的两个多声源声音检测评分场景进行评估，借此可以分析这些方案是设计用于执行细粒度时间分割、粗粒度时间分割，还是在所提出的场景中具有通用性。我们研究了参与者提出的解决方案，以分析它们对不同目标与非目标信噪比以及目标声音事件时间定位的鲁棒性。最后还提出了一项实验，以研究非目标事件对系统输出的影响。结果表明，适应于提供粗粒度分割输出的系统对不同目标与非目标信噪比具有更强的鲁棒性，并且借助特定的数据增强方法，它们对原始事件的时间定位也更具鲁棒性。最后一项实验的结果显示，当存在非目标事件时，系统往往会错误地预测出短事件。对于专为细粒度分割而定制的系统而言，情况尤其如此。
英文摘要：
This paper proposes a benchmark of submissions to Detection and Classification Acoustic Scene and Events 2021 Challenge (DCASE) Task 4 representing a sampling of the state-of-the-art in Sound Event Detection task. The submissions are evaluated according to the two polyphonic sound detection score scenarios proposed for the DCASE 2021 Challenge Task 4, which allow to make an analysis on whether submissions are designed to perform fine-grained temporal segmentation, coarse-grained temporal segmentation, or have been designed to be polyvalent on the scenarios proposed. We study the solutions proposed by participants to analyze their robustness to varying level target to non-target signal-to-noise ratio and to temporal localization of target sound events. A last experiment is proposed in order to study the impact of non-target events on systems outputs. Results show that systems adapted to provide coarse segmentation outputs are more robust to different target to non-target signal-to-noise ratio and, with the help of specific data augmentation methods, they are more robust to time localization of the original event. Results of the last experiment display that systems tend to spuriously predict short events when non-target events are present. This is particularly true for systems that are tailored to have a fine segmentation.
Simon Fraser University（西蒙菲莎大学）；Academia Sinica（中央研究院）；Dartmouth College（达特茅斯学院）
Whether literally or suggestively, the concept of soundscape is alluded in both modern and ancient music. In this study, we examine whether we can analyze and compare Western and Chinese classical music based on soundscape models. We addressed this question through a comparative study. Specifically, corpora of Western classical music excerpts (WCMED) and Chinese classical music excerpts (CCMED) were curated and annotated with emotional valence and arousal through a crowdsourcing experiment. We used a sound event detection (SED) and soundscape emotion recognition (SER) models with transfer lear…
The impact of non-target events in synthetic soundscapes for sound event detection
Université de Lorraine（洛林大学）；CNRS（法国国家科学研究中心）；Inria（法国国家信息与自动化研究所）；Loria（洛林计算机科学实验室）；Department of Information Engineering, Universita Politecnica delle Marche（马尔凯理工大学信息工程系）
ARAUS: A Large-Scale Dataset and Baseline Models of Affective Responses to Augmented Urban Soundscapes
School of Electrical and Electronic Engineering, Nanyang Technological University (NTU)（南洋理工大学电气与电子工程学院）；Center for Music Technology, Georgia Institute of Technology（佐治亚理工学院音乐技术中心）；Chungnam National University（忠南大学）
Spatial Scaper: A Library to Simulate and Augment Soundscapes for Sound Event Localization and Detection in Realistic Rooms
Music and Audio Research Laboratory, New York University（纽约大学音乐与音频研究实验室）；Viterbi School of Engineering, University of Southern California（南加州大学维特比工程学院）
We investigate an method for quantifying city characteristics based on impressions of a sound environment. The quantification of the city characteristics will be beneficial to government policy planning, tourism projects, etc. In this study, we try to predict two soundscape impressions, meaning pleasantness and eventfulness, using sound data collected by the cloud-sensing method. The collected sounds comprise meta information of recording location using Global Positioning System. Furthermore, the soundscape impressions and sound-source features are separately assigned to the cloud-sensing soun…
We propose a benchmark of state-of-the-art sound event detection systems (SED). We designed synthetic evaluation sets to focus on specific sound event detection challenges. We analyze the performance of the submissions to DCASE 2021 task 4 depending on time related modifications (time position of an event and length of clips) and we study the impact of non-target sound events and reverberation. We show that the localization in time of sound events is still a problem for SED systems. We also show that reverberation and non-target sound events are severely degrading the performance of the SED sy…
The University of Texas at Austin（德克萨斯大学奥斯汀分校）；University of Tennessee, Knoxville（田纳西大学基洛纳分校）；University of South Carolina（南卡罗来纳大学）；Arizona State University（亚利桑那州立大学）；University of Canterbury（坎特伯雷大学）；Texas A&M University（德克萨斯A&M大学）；University of Wisconsin-Madison（威斯康星大学麦迪逊分校）
New York University（纽约大学）；University of Illinois at Chicago（伊利诺伊大学芝加哥分校）；Universidade Federal Fluminense（弗鲁米嫩塞联邦大学）；New Jersey Institute of Technology（新泽西理工学院）
Noise is one of the primary quality-of-life issues in urban environments. In addition to annoyance, noise negatively impacts public health and educational performance. While low-cost sensors can be deployed to monitor ambient noise levels at high temporal resolutions, the amount of data they produce and the complexity of these data pose significant analytical challenges. One way to address these challenges is through machine listening techniques, which are used to extract features in attempts to classify the source of noise and understand temporal patterns of a city's noise situation. However,…
Misophonia Trigger Sound Detection on Synthetic Soundscapes Using a Hybrid Model with a Frozen Pre-Trained CNN and a Time-Series Module
Department of Computer Science, Nagoya Institute of Technology（名古屋技术大学计算机科学系）；International Research Center for Neurointelligence, The University of Tokyo（东京大学神经智能国际研究中心）
Training a sound event detection algorithm on a heterogeneous dataset including both recorded and synthetic soundscapes that can have various labeling granularity is a non-trivial task that can lead to systems requiring several technical choices. These technical choices are often passed from one system to another without being questioned. We propose to perform a detailed analysis of DCASE 2020 task 4 sound event detection baseline with regards to several aspects such as the type of data used for training, the parameters of the mean-teacher or the transformations applied while generating the sy…
Time-frequency localization of bird calls in dense soundscapes
Acoustic Research Laboratory, National University of Singapore（新加坡国立大学声学研究实验室）；Tropical Marine Science Institute, National University of Singapore（新加坡国立大学热带海洋科学研究所）；School of Marine Science and Technology, Northwestern Polytechnical University（西北工业大学航海学院）
This paper introduces a novel dataset for polyphonic sound event detection in urban sound monitoring use-cases. Based on isolated sounds taken from the FSD50k dataset, 20,000 polyphonic soundscapes are synthesized with sounds being randomly positioned in the stereo panorama using different loudness levels. The paper gives a detailed discussion of possible application scenarios, explains the dataset generation process in detail, and discusses current limitations of the proposed USM-SED dataset.
Generalization in birdsong classification: impact of transfer learning methods and dataset characteristics
Naturalis Biodiversity Center（自然生物多样性中心）；Xeno-Canto Foundation（异音基金会）；Landesbund fur Vogel- und Naturschutz（鸟类与自然保护州联合会）；Department of Cognitive Science and Artificial Intelligence, Tilburg University（蒂尔堡大学认知科学与人工智能系）
University of Texas at Dallas（德克萨斯大学达拉斯分校）；NEC Laboratories America（NEC美国实验室）
Sound Event Early Detection (SEED) is an essential task in recognizing the acoustic environments and soundscapes. However, most of the existing methods focus on the offline sound event detection, which suffers from the over-confidence issue of early-stage event detection and usually yield unreliable results. To solve the problem, we propose a novel Polyphonic Evidential Neural Network (PENet) to model the evidential uncertainty of the class probability with Beta distribution. Specifically, we use a Beta distribution to model the distribution of class probabilities, and the evidential uncertain…
Soundscape Captioning using Sound Affective Quality Network and Large Language Model
Machine Learning Group, University of Oxford（牛津大学机器学习组）；AIRO-IDLab, Ghent University-Imec（根特大学-imec AIRO-IDLab）；CVSSP, University of Surrey（萨里大学计算机视觉与模式识别中心）；University College London（伦敦大学学院）；WAVES Group, Ghent University（根特大学WAVES组）
Most of the existing isolated sound event datasets comprise a small number of sound event classes, usually 10 to 15, restricted to a small domain, such as domestic and urban sound events. In this work, we introduce GISE-51, a dataset spanning 51 isolated sound events belonging to a broad domain of event types. We also release GISE-51-Mixtures, a dataset of 5-second soundscapes with hard-labelled event boundaries synthesized from GISE-51 isolated sound events. We conduct baseline sound event recognition (SER) experiments on the GISE-51-Mixtures dataset, benchmarking prominent convolutional neur…
University of Augsburg（奥格斯堡大学）；Imperial College London（帝国理工学院）
The field of sound healing includes ancient practices coming from a broad range of cultures. Across such practices there is a variety of acoustic instrumentation utilised. Practitioners suggest that sound has the ability to target both mental and even physical health issues, e.g., chronic-stress, or joint-pain. Instruments including the Tibetan singing bowl and vocal chanting, are still widely used today. With the noise-floor of modern urban soundscapes continually increasing and known to impact wellbeing, methods to improve this are needed. With that in mind, this study presents the Acoustic …
NVIDIA；University of Maryland, College Park（马里兰大学 College Park 分校）；University of Science and Technology of China（中国科学技术大学）；Seoul National University（首尔国立大学）；Adobe
Automating Urban Soundscape Enhancements with AI: In-situ Assessment of Quality and Restorativeness in Traffic-Exposed Residential Areas
School of Electrical and Electronic Engineering, Nanyang Technological University（南洋理工大学电气与电子工程学院）；Center for Music Technology, Georgia Institute of Technology（佐治亚理工学院音乐技术中心）；Building & Research Institute, Housing & Development Board（建屋发展局建筑与研究院）；Department of Architectural Engineering, Chungnam National University（忠南大学建筑工程系）；UCL Institute for Environmental Design and Engineering, The Bartlett, University College London（伦敦大学学院巴特lett环境设计与工程学院）；Nanyang Business School, Nanyang Technological University（南洋理工大学南洋商学院）
Mid Sweden University（瑞典中部大学）；Ton Duc Thang University（孙德胜大学）
Monitoring the conditions of machines is vital in the manufacturing industry. Early detection of faulty components in machines for stopping and repairing the failed components can minimize the downtime of the machine. This article presents an approach to detect the failure occurring in drill machines based on drill sounds from Valmet AB. The drill dataset includes three classes: anomalous sounds, normal sounds, and irrelevant sounds, which are also labeled as "Broken", "Normal", and "Other", respectively. Detecting drill failure effectively remains a challenge due to the following reasons. The…
Generating Diverse Audio-Visual 360 Soundscapes for Sound Event Localization and Detection
University of Southern California（南加州大学）；Universidad Nacional Autónoma de México（墨西哥国立自治大学）；Universidad Autónoma de Tamaulipas（塔毛利帕斯自治大学）；Queen Mary University of London（伦敦大学玛丽皇后学院）
New York University（纽约大学）；New Jersey Institute of Technology（新泽西理工学院）；Northwestern University（西北大学）
While the estimation of what sound sources are, when they occur, and from where they originate has been well-studied, the estimation of how loud these sound sources are has been often overlooked. Current solutions to this task, which we refer to as source-specific sound level estimation (SSSLE), suffer from challenges due to the impracticality of acquiring realistic data and a lack of robustness to realistic recording conditions. Recently proposed weakly supervised source separation offer a means of leveraging clip-level source annotations to train source separation models, which we augment wi…
Joint Prediction of Audio Event and Annoyance Rating in an Urban Soundscape by Hierarchical Graph Representation Learning
Ghent University（根特大学）；University of Leicester（莱斯特大学）；Shenzhen University（深圳大学）；University College London（伦敦大学学院）；Ghent University-Imec（根特大学-imec）；University of Surrey（萨里大学）
Tampere University（坦佩雷大学）；Sony Group Corporation（索尼集团公司）
This report presents the Sony-TAu Realistic Spatial Soundscapes 2022 (STARS22) dataset for sound event localization and detection, comprised of spatial recordings of real scenes collected in various interiors of two different sites. The dataset is captured with a high resolution spherical microphone array and delivered in two 4-channel formats, first-order Ambisonics and tetrahedral microphone array. Sound events in the dataset belonging to 13 target sound classes are annotated both temporally and spatially through a combination of human annotation and optical tracking. The dataset serves as t…
University of Chinese Academy of Sciences（中国科学院大学）；Xi'an Jiaotong Liverpool University（西交利物浦大学）；ByteDance（字节跳动）；University of Surrey（萨里大学）
Sound event localization and detection (SELD) is a joint task of sound event detection and direction-of-arrival estimation. In DCASE 2022 Task 3, types of data transform from computationally generated spatial recordings to recordings of real-sound scenes. Our system submitted to the DCASE 2022 Task 3 is based on our previous proposed Event-Independent Network V2 (EINV2) with a novel data augmentation method. Our method employs EINV2 with a track-wise output format, permutation-invariant training, and a soft parameter-sharing strategy, to detect different sound events of the same class but in d…
Spike Encoding for Environmental Sound: A Comparative Benchmark
Audio and Neuromorphic Processing Group, Instituto Tecnologico de Informatica (ITI), Valencia, Spain（音频与神经形态处理组，技术信息研究所（ITI），西班牙瓦伦西亚）；Department of Computer Science, Universitat de Valencia, 46100 Burjassot, Spain（计算机科学系，瓦伦西亚大学，西班牙布尔贾索特）
Environmental Sound Deepfake Detection Challenge: An Overview
1 School of Electrical Engineering, KAIST, Daejeon, Republic of Korea 2 University of Melbourne, Australia, 3 Fortemedia Singapore, Singapore, 4 Xi’an University of Posts \& Telecommunications, Xi’an, China, 5 Xi'an Lianfeng Acoustic Technologies Co., Ltd., China
SEABAD: A Tropical Bird Activity Detection Dataset for Passive Acoustic Monitoring
Faculty of Computer Science and Information Technology, Universiti Malaya（马来亚大学计算机科学与信息技术学院）；Faculty of Electrical Engineering, Universiti Teknologi Malaysia（马来西亚科技大学电气工程学院）
Institute of Computing Technology, Chinese Academy of Sciences（中国科学院计算技术研究所）；University of Chinese Academy of Sciences（中国科学院大学）
Due to the limitation of strong-labeled sound event detection data set, using synthetic data to improve the sound event detection system performance has been a new research focus. In this paper, we try to exploit the usage of synthetic data to improve the feature representation. Based on metric learning, we proposed inter-frame distance loss function for domain adaptation, and prove the effectiveness of it on sound event detection. We also applied multi-task learning with synthetic data. We find the the best performance can be achieved when the two methods being used together. The experiment o…
ESDD 2026: Environmental Sound Deepfake Detection Challenge Evaluation Plan
1 School of Electrical Engineering, KAIST, Daejeon, Republic of Korea 2 University of Melbourne, Australia 3 Fortemedia Singapore, Singapore 4 Xi'an University of Posts \& Telecommunications, Xi'an, China
The First Environmental Sound Deepfake Detection Challenge: Benchmarking Robustness, Evaluation, and Insights
School of Electrical Engineering, KAIST, Daejeon, Republic of Korea（韩国成均馆大学电气工程学院）；University of Melbourne, Australia（墨尔本大学）；Fortemedia Singapore, Singapore（新加坡Fortemedia公司）；Xi’an University of Posts & Telecommunications, Xi’an, China（西安邮电大学）；Xi'an Lianfeng Acoustic Technologies Co., Ltd., China（西安联丰声学技术有限公司）
