DCASE 2023 Challenge Task 2 Development Dataset
收藏资源简介：
<strong>Description</strong> This dataset is the "development dataset" for the DCASE 2023 Challenge Task 2 "First-Shot Unsupervised Anomalous Sound Detection for Machine Condition Monitoring". The data consists of the normal/anomalous operating sounds of seven types of real/toy machines. Each recording is a single-channel 10-second audio that includes both a machine's operating sound and environmental noise. The following seven types of real/toy machines are used in this task: ToyCar ToyTrain Fan Gearbox Bearing Slide rail Valve <strong>Overview of the task</strong> <strong>Anomalous sound detection (ASD) is the task of identifying whether the sound emitted from a target machine is normal or anomalous. </strong>Automatic detection of mechanical failure is an essential technology in the fourth industrial revolution, which involves artificial-intelligence-based factory automation. Prompt detection of machine anomalies by observing sounds is useful for monitoring the condition of machines. This task is the follow-up from DCASE 2020 Task 2 to DCASE 2022 Task 2. The task this year is to develop an ASD system that meets the following four requirements. <strong>1. Train a model using only normal sound (unsupervised learning scenario)</strong> Because anomalies rarely occur and are highly diverse in real-world factories, it can be difficult to collect exhaustive patterns of anomalous sounds. Therefore, the system must detect unknown types of anomalous sounds that are not provided in the training data. This is the same requirement as in the previous tasks. <strong>2. Detect anomalies regardless of domain shifts (domain generalization task) </strong> In real-world cases, the operational states of a machine or the environmental noise can change to cause domain shifts. Domain-generalization techniques can be useful for handling domain shifts that occur frequently or are hard-to-notice. In this task, the system is required to use domain-generalization techniques for handling these domain shifts. This requirement is the same as in DCASE 2022 Task 2. <strong>3. Train a model for a completely new machine type</strong> For a completely new machine type, hyperparameters of the trained model cannot be tuned. Therefore, the system should have the ability to train models without additional hyperparameter tuning. <strong>4. Train a model using only one machine from its machine type</strong> While sounds from multiple machines of the same machine type can be used to enhance detection performance, it is often the case that sound data from only one machine are available for a machine type. In such a case, the system should be able to train models using only one machine from a machine type. The last two requirements are newly introduced in DCASE 2023 Task2 as the "first-shot problem". <strong>Definition</strong> We first define key terms in this task: "machine type," "section," "source domain," "target domain," and "attributes.". "Machine type" indicates the type of machine, which in the development dataset is one of seven: fan, gearbox, bearing, slide rail, valve, ToyCar, and ToyTrain. A section is defined as a subset of the dataset for calculating performance metrics. The source domain is the domain under which most of the training data and some of the test data were recorded, and the target domain is a different set of domains under which some of the training data and some of the test data were recorded. There are differences between the source and target domains in terms of operating speed, machine load, viscosity, heating temperature, type of environmental noise, signal-to-noise ratio, etc. Attributes are parameters that define states of machines or types of noise. <strong>Dataset</strong> This dataset consists of seven machine types. For each machine type, one section is provided, and the section is a complete set of training and test data. For each section, this dataset provides (i) 990 clips of normal sounds in the source domain for training, (ii) ten clips of normal sounds in the target domain for training, and (iii) 100 clips each of normal and anomalous sounds for the test. The source/target domain of each sample is provided. Additionally, the attributes of each sample in the training and test data are provided in the file names and attribute csv files. <strong>File names and attribute csv files</strong> File names and attribute csv files provide reference labels for each clip. The given reference labels for each training/test clip include machine type, section index, normal/anomaly information, and attributes regarding the condition other than normal/anomaly. The machine type is given by the directory name. The section index is given by their respective file names. For the datasets other than the evaluation dataset, the normal/anomaly information and the attributes are given by their respective file names. Attribute csv files are for easy access to attributes that cause domain shifts. In these files, the file names, name of parameters that cause domain shifts (domain shift parameter, dp), and the value or type of these parameters (domain shift value, dv) are listed. Each row takes the following format: [filename (string)], [d1p (string)], [d1v (int | float | string)], [d2p], [d2v]... <strong>Recording procedure</strong> Normal/anomalous operating sounds of machines and its related equipment are recorded. Anomalous sounds were collected by deliberately damaging target machines. For simplifying the task, we use only the first channel of multi-channel recordings; all recordings are regarded as single-channel recordings of a fixed microphone. We mixed a target machine sound with environmental noise, and only noisy recordings are provided as training/test data. The environmental noise samples were recorded in several real factory environments. We will publish papers on the dataset to explain the details of the recording procedure by the submission deadline. <strong>Directory structure</strong> - /dev_data - /raw<br> - /fan<br> - /train (only normal clips) <br> - /section_00_source_train_normal_0000_<attribute>.wav <br> - ... <br> - /section_00_source_train_normal_0989_<attribute>.wav <br> - /section_00_target_train_normal_0000_<attribute>.wav <br> - ... <br> - /section_00_target_train_normal_0009_<attribute>.wav <br> - /test <br> - /section_00_source_test_normal_0000_<attribute>.wav <br> - ... <br> - /section_00_source_test_normal_0049_<attribute>.wav <br> - /section_00_source_test_anomaly_0000_<attribute>.wav <br> - ... <br> - /section_00_source_test_anomaly_0049_<attribute>.wav <br> - /section_00_target_test_normal_0000_<attribute>.wav<br> - ... <br> - /section_00_target_test_normal_0049_<attribute>.wav <br> - /section_00_target_test_anomaly_0000_<attribute>.wav <br> - ... <br> - /section_00_target_test_anomaly_0049_<attribute>.wav <br> - attributes_00.csv (attribute csv for section 00)<br> - /gearbox (The other machine types have the same directory structure as fan.) <br> - /bearing<br> - /slider (`slider` means "slide rail")<br> - /ToyCar <br> - /ToyTrain <br> - /valve <strong>Baseline system</strong> The baseline system is available on the Github repository dcase2023_task2_baseline_ae.The baseline systems provide a simple entry-level approach that gives a reasonable performance in the dataset of Task 2. They are good starting points, especially for entry-level researchers who want to get familiar with the anomalous-sound-detection task. <strong>Condition of use</strong> This dataset was created jointly by <strong>Hitachi, Ltd. </strong>and <strong>NTT Corporation</strong> and is available under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0) license. <strong>Citation</strong> If you use this dataset, please cite all the following papers. We will publish a paper on the description of the DCASE 2023 Task 2, so pleasure make sure to cite the paper, too. Noboru Harada, Daisuke Niizumi, Yasunori Ohishi, Daiki Takeuchi, and Masahiro Yasuda. <em>First-shot anomaly detection for machine condition monitoring: A domain generalization baseline. In arXiv e-prints: 2303.00455</em>, 2023. [URL] Kota Dohi, Tomoya Nishida, Harsh Purohit, Ryo Tanabe, Takashi Endo, Masaaki Yamamoto, Yuki Nikaido, and Yohei Kawaguchi. <em>MIMII DG: sound dataset for malfunctioning industrial machine investigation and inspection for domain generalization task.</em> In Proceedings of the 7th Detection and Classification of Acoustic Scenes and Events 2022 Workshop (DCASE2022), 31-35. Nancy, France, November 2022, . [URL] Noboru Harada, Daisuke Niizumi, Daiki Takeuchi, Yasunori Ohishi, Masahiro Yasuda, and Shoichiro Saito. <em>ToyADMOS2: another dataset of miniature-machine operating sounds for anomalous sound detection under domain shift conditions.</em> In Proceedings of the 6th Detection and Classification of Acoustic Scenes and Events 2021 Workshop (DCASE2021), 1–5. Barcelona, Spain, November 2021. [URL] <strong>Contact</strong> If there is any problem, please contact us: Kota Dohi, kota.dohi.gr@hitachi.com Keisuke Imoto, keisuke.imoto@ieee.org Noboru Harada, noboru@ieee.org Daisuke Niizumi, daisuke.niizumi.dt@hco.ntt.co.jp Yohei Kawaguchi, yohei.kawaguchi.xk@hitachi.com
<strong>数据集说明</strong> 本数据集为DCASE 2023挑战赛任务2「面向机器状态监测的首样本无监督异常声音检测」的开发集（development dataset）。该数据集包含7类实体/玩具机器的正常与异常运行音频。每条录音均为单通道10秒音频，同时包含机器运行声与环境噪声。本任务使用的7类实体/玩具机器分别为：玩具车（ToyCar）、玩具火车（ToyTrain）、风扇（Fan）、齿轮箱（Gearbox）、轴承（Bearing）、滑轨（Slide rail）、阀门（Valve）。<strong>任务概述</strong> <strong>异常声音检测（Anomalous Sound Detection, ASD）</strong>指识别目标机器发出的声音是否属于正常或异常状态。机械故障自动检测是第四次工业革命的核心技术之一，其依托人工智能实现工厂自动化。通过声音监测实现机器异常的快速检测，对于机器状态监测具有重要应用价值。本任务是DCASE 2020任务2至DCASE 2022任务2的延续。今年的任务目标是开发满足以下四项要求的异常声音检测系统。<strong>1. 仅使用正常声音训练模型（无监督学习场景）</strong> 在现实工厂中，异常样本极少出现且类型多样，难以收集全覆盖的异常声音模式。因此，系统需能够检测训练集中未出现过的未知类型异常声音，这与往届任务的要求一致。<strong>2. 无视域偏移检测异常（域泛化任务）</strong> 在实际场景中，机器运行状态或环境噪声的变化可能引发域偏移。域泛化（domain generalization）技术可有效处理频繁出现或难以察觉的域偏移问题。本任务要求系统采用域泛化技术应对此类域偏移，这与DCASE 2022任务2的要求一致。<strong>3. 为全新机器类型训练模型</strong> 对于全新的机器类型，无法对训练模型的超参数进行调优。因此，系统需具备无需额外超参数调优即可完成模型训练的能力。<strong>4. 仅使用单台同类型机器训练模型</strong> 尽管可通过同类型多台机器的音频数据提升检测性能，但实际场景中常仅能获取单台机器的音频数据。在此情况下，系统需能够仅基于单台同类型机器的音频完成模型训练。后两项要求为DCASE 2023任务2新增的「首样本问题（first-shot problem）」。<strong>术语定义</strong> 首先明确本任务中的关键术语：「机器类型」、「区段（section）」、「源域」、「目标域」与「属性」。「机器类型」指机器的类别，本开发集中的机器类型共7类：风扇（Fan）、齿轮箱（Gearbox）、轴承（Bearing）、滑轨（Slide rail）、阀门（Valve）、玩具车（ToyCar）与玩具火车（ToyTrain）。区段指用于计算性能指标的数据集子集。源域指大部分训练数据与部分测试数据的录制域，目标域指另一组包含部分训练与测试数据的录制域。源域与目标域在运行速度、机器负载、粘度、加热温度、环境噪声类型、信噪比等参数上存在差异。属性指定义机器状态或噪声类型的参数。<strong>数据集构成</strong> 本数据集包含7类机器。每类机器对应一个区段，该区段包含完整的训练与测试数据。每个区段的数据分为三部分：(i) 990条源域正常声音训练片段，(ii) 10条目标域正常声音训练片段，(iii) 测试集各包含100条正常与异常声音片段。每个样本的源/目标域信息均已提供。此外，训练与测试集中每个样本的属性信息已通过文件名与属性CSV文件给出。<strong>文件名与属性CSV文件</strong> 文件名与属性CSV文件为每条音频片段提供参考标签。每条训练/测试片段的参考标签包括机器类型、区段索引、正常/异常信息，以及除正常/异常外的状态属性。机器类型由所在目录名给出，区段索引由文件名给出。除评估集外的数据集，其正常/异常信息与属性均包含在文件名中。属性CSV文件用于便捷获取引发域偏移的属性信息，文件中列出了文件名、引发域偏移的参数名称（域偏移参数，domain shift parameter, dp），以及这些参数的取值或类型（域偏移值，domain shift value, dv）。每行格式如下：[文件名（字符串）], [d1p（字符串）], [d1v（整数|浮点数|字符串）], [d2p], [d2v]……<strong>录制流程</strong> 录制机器及其配套设备的正常与异常运行声音。异常声音通过人为损坏目标机器获取。为简化任务，本数据集仅使用多通道录音的第一声道，所有录音均视为固定麦克风采集的单通道音频。将目标机器声音与环境噪声混合，仅提供带噪的训练与测试音频。环境噪声样本录制自多个真实工厂环境。我们将在提交截止日前发布论文详细阐述本数据集的录制流程细节。<strong>目录结构</strong> - /dev_data - /raw<br> - /fan<br> - /train（仅包含正常片段） <br> - /section_00_source_train_normal_0000_<attribute>.wav <br> - …… <br> - /section_00_source_train_normal_0989_<attribute>.wav <br> - /section_00_target_train_normal_0000_<attribute>.wav <br> - …… <br> - /section_00_target_train_normal_0009_<attribute>.wav <br> - /test <br> - /section_00_source_test_normal_0000_<attribute>.wav <br> - …… <br> - /section_00_source_test_normal_0049_<attribute>.wav <br> - /section_00_source_test_anomaly_0000_<attribute>.wav <br> - …… <br> - /section_00_source_test_anomaly_0049_<attribute>.wav <br> - /section_00_target_test_normal_0000_<attribute>.wav<br> - …… <br> - /section_00_target_test_normal_0049_<attribute>.wav <br> - /section_00_target_test_anomaly_0000_<attribute>.wav <br> - …… <br> - /section_00_target_test_anomaly_0049_<attribute>.wav <br> - attributes_00.csv（区段00的属性CSV文件）<br> - /gearbox（其余机器类型的目录结构与风扇一致） <br> - /bearing<br> - /slider（此处「slider」指代「滑轨（Slide rail）」）<br> - /ToyCar <br> - /ToyTrain <br> - /valve<strong>基线系统</strong> 基线系统可在Github仓库dcase2023_task2_baseline_ae中获取。基线系统提供了简单的入门级方案，可在本任务2的数据集上获得合理的检测性能，尤其适合希望熟悉异常声音检测任务的入门研究者作为研究起点。<strong>使用条款</strong> 本数据集由日立有限公司（Hitachi, Ltd.）与日本电报电话公司（NTT Corporation）联合制作，采用知识共享署名-非商业性使用-相同方式共享4.0国际许可协议（Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International, CC BY-NC-SA 4.0）进行发布。<strong>引用要求</strong> 若使用本数据集，请引用以下全部论文。我们将在提交截止日前发布关于DCASE 2023任务2的详细说明论文，请务必一并引用该论文。<br>Noboru Harada, Daisuke Niizumi, Yasunori Ohishi, Daiki Takeuchi, and Masahiro Yasuda. <em>首样本机器状态监测异常检测：一种域泛化基线方法</em>，arXiv预印本：2303.00455，2023. [URL]<br>Kota Dohi, Tomoya Nishida, Harsh Purohit, Ryo Tanabe, Takashi Endo, Masaaki Yamamoto, Yuki Nikaido, and Yohei Kawaguchi. <em>MIMII DG：面向域泛化任务的工业机器故障检测与检查音频数据集</em>，收录于第7届声学场景检测与分类事件研讨会（DCASE2022）论文集，31-35. 法国南希，2022年11月. [URL]<br>Noboru Harada, Daisuke Niizumi, Daiki Takeuchi, Yasunori Ohishi, Masahiro Yasuda, and Shoichiro Saito. <em>ToyADMOS2：面向域偏移条件下异常声音检测的微型机器运行音频数据集</em>，收录于第6届声学场景检测与分类事件研讨会（DCASE2021）论文集，1-5. 西班牙巴塞罗那，2021年11月. [URL]<strong>联系方式</strong> 如有任何问题，请联系以下人员：<br>Kota Dohi, kota.dohi.gr@hitachi.com<br>Keisuke Imoto, keisuke.imoto@ieee.org<br>Noboru Harada, noboru@ieee.org<br>Daisuke Niizumi, daisuke.niizumi.dt@hco.ntt.co.jp<br>Yohei Kawaguchi, yohei.kawaguchi.xk@hitachi.com
