Unsupervised Detection of Anomalous Sound Based on Deep Learning and the Neyman–Pearson Lemma

Koizumi, Yohei; Saito, Shoichiro; Uematsu, Hitoshi; Kawachi, Yuta; Harada, Noboru

doi:10.1109/taslp.2018.2877258

Cited by 135 publications

(78 citation statements)

References 25 publications

(46 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…One characteristic of AED task is the ratio of concerned acoustic event to the background is quite small, which is similar to the anomaly event detection task [32], [33]. The difference is that we already know the types of audio events in AED tasks [34], so we need to look for events with known features, but in anomaly event detection we have no information about the abnormal samples to be discriminated. For anomaly event detection task, [34] proposes a framework: firstly, they construct a complete set in a manifold space, then subtract the normal part from the complete set to obtain information about anomalous audio and recognize it.…”

Section: Anomaly Event Detectionmentioning

confidence: 99%

“…The difference is that we already know the types of audio events in AED tasks [34], so we need to look for events with known features, but in anomaly event detection we have no information about the abnormal samples to be discriminated. For anomaly event detection task, [34] proposes a framework: firstly, they construct a complete set in a manifold space, then subtract the normal part from the complete set to obtain information about anomalous audio and recognize it. Furthermore, [35] uses the Kullback-Leibler (KL) divergence to measure the similarity between normal and abnormal samples according to the short-time Fourier transform (STFT) feature.…”

Section: Anomaly Event Detectionmentioning

confidence: 99%

See 1 more Smart Citation

Adaptive Multi-Scale Detection of Acoustic Events

Ding

2020

IEEE/ACM Trans. Audio Speech Lang. Process.

View full text Add to dashboard Cite

The goal of acoustic (or sound) events detection (AED or SED) is to predict the temporal position of target events in given audio segments. This task plays a significant role in safety monitoring, acoustic early warning and other scenarios. However, the deficiency of data and diversity of acoustic event sources make the AED task a tough issue, especially for prevalent data-driven methods. In this paper, we start from analyzing acoustic events according to their time-frequency domain properties, showing that different acoustic events have different time-frequency scale characteristics. Inspired by the analysis, we propose an adaptive multi-scale detection (AdaMD) method. By taking advantage of hourglass neural network and gated recurrent unit (GRU) module, our AdaMD produces multiple predictions at different temporal and frequency resolutions. An adaptive training algorithm is subsequently adopted to combine multi-scale predictions to enhance the overall capability. Experimental results on Detection and Classification of Acoustic Scenes and Events 2017 (DCASE 2017) Task 2, DCASE 2016 Task 3 and DCASE 2017 Task 3 demonstrate that the AdaMD outperforms published state-of-the-art competitors in terms of the metrics of event error rate (ER) and F1-score. The verification experiment on our collected factory mechanical dataset also proves the noiseresistant capability of the AdaMD, providing the possibility for it to be deployed in the complex environment. Index Terms-rare acoustic event detection, adaptive multiscale, hourglass network PLACE PHOTO HERE Liang He Liang HE received the B.S. degree in communication engineering from Civil Aviation

show abstract

Section: Anomaly Event Detectionmentioning

confidence: 99%

Section: Anomaly Event Detectionmentioning

confidence: 99%

Adaptive Multi-Scale Detection of Acoustic Events

Ding

2020

IEEE/ACM Trans. Audio Speech Lang. Process.

View full text Add to dashboard Cite

show abstract

“…This results in the anomalous sounds also being reconstructed and having small A θ (xt). To increase A θ (xt) for anomalous sounds, we previously proposed a training method of an AE that works to increase A θ (xt) of simulated anomalous sounds by defining the anomalous sound as "non-normal" [11]. As a simplified implementation of our method [11], a mini-batch of anomalous sounds {x (a) j } Ma j=1 can be generated by adding sounds of other things (hereinafter, something-else sounds) as…”

Section: Dnn-based Unsupervised-adsmentioning

confidence: 99%

Batch Uniformization for Minimizing Maximum Anomaly Score of Dnn-Based Anomaly Detection in Sounds

Saito

Yamaguchi

Murata

et al. 2019

2019 IEEE Workshop on Applications of Signal Processing to Audio and Acoustics (WASPAA)

Self Cite

View full text Add to dashboard Cite

Use of an autoencoder (AE) as a normal model is a state-of-theart technique for unsupervised-anomaly detection in sounds (ADS). The AE is trained to minimize the sample mean of the anomaly score of normal sounds in a mini-batch. One problem with this approach is that the anomaly score of rare-normal sounds becomes higher than that of frequent-normal sounds, because the sample mean is strongly affected by frequent-normal samples, resulting in preferentially decreasing the anomaly score of frequent-normal samples. To decrease anomaly scores for both frequent-and rarenormal sounds, we propose batch uniformization, a training method for unsupervised-ADS for minimizing a weighted average of the anomaly score on each sample in a mini-batch. We used the reciprocal of the probabilistic density of each sample as the weight, more intuitively, a large weight is given for rare-normal sounds. Such a weight works to give a constant anomaly score for both frequent-and rare-normal sounds. Since the probabilistic density is unknown, we estimate it by using the kernel density estimation on each training mini-batch. Verification-and objective-experiments show that the proposed batch uniformization improves the performance of unsupervised-ADS.Index Terms-Anomaly detection in sounds, uniform distribution, kernel density estimation and deep learning.

show abstract

“…Examples include vibration sensorbased approaches [1][2][3][4], temperature sensor-based approaches [5], and pressure sensor-based approaches [6]. Another approach is to detect anomalies from sound by using technologies for acoustic scene classification and event detection [7][8][9][10][11][12][13]. Remarkable advancements have been made in the classification of acoustic scenes and the detection of acoustic events, and there are many promising state-of-the-art studies in this vein [14][15][16].…”

Section: Introductionmentioning

confidence: 99%

MIMII Dataset: Sound Dataset for Malfunctioning Industrial Machine Investigation and Inspection

Purohit¹,

Tanabe²,

Ichige³

et al. 2019

Proceedings of the Detection and Classification of Acoustic Scenes And Events 2019 Workshop (DCASE2019)

223

View full text Add to dashboard Cite

Factory machinery is prone to failure or breakdown, resulting in significant expenses for companies. Hence, there is a rising interest in machine monitoring using different sensors including microphones. In the scientific community, the emergence of public datasets has led to advancements in acoustic detection and classification of scenes and events, but there are no public datasets that focus on the sound of industrial machines under normal and anomalous operating conditions in real factory environments. In this paper, we present a new dataset of industrial machine sounds that we call a sound dataset for malfunctioning industrial machine investigation and inspection (MIMII dataset). Normal sounds were recorded for different types of industrial machines (i.e., valves, pumps, fans, and slide rails), and to resemble a real-life scenario, various anomalous sounds were recorded (e.g., contamination, leakage, rotating unbalance, and rail damage). The purpose of releasing the MIMII dataset is to assist the machine-learning and signalprocessing community with their development of automated facility maintenance.

show abstract

Unsupervised Detection of Anomalous Sound Based on Deep Learning and the Neyman–Pearson Lemma

Cited by 135 publications

References 25 publications

Adaptive Multi-Scale Detection of Acoustic Events

Adaptive Multi-Scale Detection of Acoustic Events

Batch Uniformization for Minimizing Maximum Anomaly Score of Dnn-Based Anomaly Detection in Sounds

MIMII Dataset: Sound Dataset for Malfunctioning Industrial Machine Investigation and Inspection

Contact Info

Product

Resources

About