Filters








13,925 Hits in 6.9 sec

Leveraging the urban soundscape: Auditory perception for smart vehicles

Letizia Marchegiani, Ingmar Posner
2017 2017 IEEE International Conference on Robotics and Automation (ICRA)  
We first model the urban soundscape and use anomaly detection to identify the presence of an anomalous sound, and later determine the nature of this sound.  ...  When compared to traditional feature representations, such as Mel-frequency cepstrum coefficients, our framework shows an improvement of up to 31% in the classification rate.  ...  Our analysis is twofold: we first detect the presence of anomalous sounds using one-class classification and then further process the detected anomalous sounds to identify their nature.  ... 
doi:10.1109/icra.2017.7989774 dblp:conf/icra/MarchegianiP17 fatcat:uq25i2yvhndg7nbof4lszwlz54

A Large-Scale Benchmark Dataset for Anomaly Detection and Rare Event Classification for Audio Forensics

Ahmed Abbasi, Abdul Rehman Javed, Amanullah Yasin, Zunera Jalil, Natalia Kryvinska, Usman Tariq
2022 IEEE Access  
., beach, restaurant, and train) to focus on both detection of anomalous audio and classification of rare sound (e.g., events-baby cry, gunshots, broken glasses, footsteps) events for audio forensics.  ...  minimum number of best-performing features for optimum performance using principal component analysis (PCA).  ...  Visual representation of proposed approach for detection of anomalous audio and classification of rare sound events. FIGURE 4 . 4 FIGURE 4. Spectrogram of audio signal. FIGURE 5 . 5 FIGURE 5.  ... 
doi:10.1109/access.2022.3166602 fatcat:2u6hcb7dvra45divf2prokn5ta

Deep Dense and Convolutional Autoencoders for Unsupervised Anomaly Detection in Machine Condition Sounds [article]

Alexandrine Ribeiro, Luis Miguel Matos, Pedro Jose Pereira, Eduardo C. Nunes, Andre L. Ferreira, Paulo Cortez, Andre Pilastri
2020 arXiv   pre-print
The challenge involves an unsupervised learning to detect anomalous sounds, thus only normal machine working condition samples are available during the training process.  ...  This technical report describes two methods that were developed for Task 2 of the DCASE 2020 challenge.  ...  CONCLUSIONS In this paper, we proposed two Autoencoder (AE) models for an unsupervised Anomalous Sound Detection (ASD), for the second task of the DCASE 2020 challenge.  ... 
arXiv:2006.10417v2 fatcat:clqtupdyh5g35am6sqzttagqxu

Detecting Drill Failure in the Small Short-sound Drill Dataset [article]

Thanh Tran, Nhat Truong Pham, Jan Lundgren
2021 arXiv   pre-print
Detecting drill failure effectively remains a challenge due to the following reasons. The waveform of drill sound is complex and short for detection.  ...  We then proposed a convolutional neural network (CNN) combined with a long short-term memory (LSTM) to extract features from log-Mel spectrograms and learn global high-level feature representation for  ...  On average, broken drill bits account for just 0.16% of the total number of drill sounds recorded (67 anomalous sounds out of 41,250 total drill sounds). Therefore, broken drills are hard to detect.  ... 
arXiv:2108.11089v2 fatcat:zzwtd45hvvfknf2fjvxdjsnvaa

Acoustic Anomaly Detection of Mechanical Failures in Noisy Real-Life Factory Environments

Yuki Tagawa, Rytis Maskeliūnas, Robertas Damaševičius
2021 Electronics  
The main element of the proposed methodology is a generative adversarial network (GAN) used for the reconstruction of sound signal reconstruction and the detection of anomalies.  ...  Anomaly detection without employing dedicated sensors for each industrial machine is recognized as one of the essential techniques for preventive maintenance and is especially important for factories with  ...  Acknowledgments: The authors express their gratitude to the creators of the MIMII dataset for making the data available for research: Harsh Purohit, Ryo Tanabe, Kenji Ichige, Takashi Endo, Yuki Nikaido  ... 
doi:10.3390/electronics10192329 fatcat:amdwz3nkhzdhxd7bsmcnyg33uq

Abnormal sound event detection using temporal trajectories mixtures

Debmalya Chakrabarty, Mounya Elhilali
2016 2016 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)  
Index Terms-Anomalous sound events, Hierarchical network, Convolutional feature representation, Mixture of temporal trajectory models 978-1-4799-9988-0/16/$31.00  ...  Detection of anomalous sound events in audio surveillance is a challenging task when applied to realistic settings.  ...  Operating on this feature analysis often comes a robust backend classifier whose role is to capture variability across different instances of the sound class.  ... 
doi:10.1109/icassp.2016.7471668 dblp:conf/icassp/ChakrabartyE16 fatcat:rwilvhm4kfbvvg7zc2yxvpfnb4

Acoustic Anomaly Detection for Machine Sounds based on Image Transfer Learning [article]

Robert Müller, Fabian Ritz, Steffen Illium, Claudia Linnhoff-Popien
2020 arXiv   pre-print
We then use these features to train a variety of anomaly detection models and show that this improves results compared to convolutional autoencoders in recordings of four different factory machines in  ...  In industrial applications, the early detection of malfunctioning factory machinery is crucial. In this paper, we consider acoustic malfunction detection via transfer learning.  ...  Conclusion In this work, we thoroughly studied acoustic anomaly detection for machine sounds.  ... 
arXiv:2006.03429v2 fatcat:comzhbsfbbeibirmotbenzxlzy

Using UMAP to Inspect Audio Data for Unsupervised Anomaly Detection under Domain-Shift Conditions [article]

Andres Fernandez, Mark D. Plumbley
2021 arXiv   pre-print
The goal of Unsupervised Anomaly Detection (UAD) is to detect anomalous signals under the condition that only non-anomalous (normal) data is available beforehand.  ...  representation and detection approaches.  ...  INTRODUCTION The goal of Unsupervised Anomaly Detection (UAD) is to detect anomalous instances under the condition that only non-anomalous (i.e. normal) instances are available beforehand.  ... 
arXiv:2107.10880v2 fatcat:ru6op4432fgobnj7qlho7euge4

A multiresolution analysis for detection of abnormal lung sounds

D. Emmanouilidou, K. Patil, J. West, M. Elhilali
2012 2012 Annual International Conference of the IEEE Engineering in Medicine and Biology Society  
Automated analysis and detection of abnormal lung sound patterns has great potential for improving access to standardized diagnosis of pulmonary diseases, especially in low-resource settings.  ...  In the current study, we develop signal processing tools for analysis of paediatric auscultations recorded under non-ideal noisy conditions.  ...  The study presents an alternative signal processing scheme and develops an analysis methodology for assessment of model accuracy for detecting adventitious sounds and distinguishing them from normal breathing  ... 
doi:10.1109/embc.2012.6346630 pmid:23366591 pmcid:PMC4087194 fatcat:2qnortjjojeqtho73xvw4momea

Semi-Supervised Machine Condition Monitoring by Learning Deep Discriminative Audio Features

Iordanis Thoidis, Marios Giouvanakis, George Papanikolaou
2021 Electronics  
In this study, we aim to learn highly descriptive representations for a wide set of machinery sounds and exploit this knowledge to perform condition monitoring of mechanical equipment.  ...  By fusing the supervised feature learning approach with an unsupervised deep one-class neural network, we are able to model the characteristics of each source and implicitly detect anomalies in different  ...  Acknowledgments: We would like to thank Lazaros Vrysis for his collaboration in the conceptual-  ... 
doi:10.3390/electronics10202471 fatcat:cqzs32cy2vhmrkxu3cg4j7izfe

Semi-supervised and Unsupervised Methods for Heart Sounds Classification in Restricted Data Environments [article]

Balagopal Unnikrishnan, Pranshu Ranjan Singh, Xulei Yang, Matthew Chin Heng Chua
2020 arXiv   pre-print
Automated heart sounds classification is a much-required diagnostic tool in the view of increasing incidences of heart related diseases worldwide.  ...  The potential of the proposed semi-supervised and unsupervised methods may lead to a workflow tool in the future for the creation of higher quality datasets.  ...  While training, the latent representation would provide a set of features which represent the training samples. These feature set would help discriminate between normal and anomalous samples.  ... 
arXiv:2006.02610v1 fatcat:ykcekjzrwbg3xmtwkwqxrwnlhy

A Squeeze-and-Excitation and Transformer based Cross-task System for Environmental Sound Recognition [article]

Jisheng Bai, Jianfeng Chen, Mou Wang, Muhammad Saad Ayub
2022 arXiv   pre-print
In this paper, we propose a cross-task system for three different tasks of ESR: acoustic scene classification, urban sound tagging, and anomalous sound detection.  ...  acoustic features.  ...  First of all, Fig. 8 shows the number of recordings of Analysis of anomalous sound detection In Sec. 5.1.4, we have shown some examples of spectrograms and analyzed the cross-task performance of different  ... 
arXiv:2203.08350v1 fatcat:fut3y5q6mjedlkvhteghyjni6e

Review of anomalous sound event detection approaches

Amirul Sadikin Md Affendi, Marina Yusoff
2019 IAES International Journal of Artificial Intelligence (IJ-AI)  
<p>This paper presents a review of anomalous sound event detection(SED) approaches.  ...  Autonomous SED could be used for early detection and prevention.  ...  ACKNOWLEDGEMENTS Universiti Teknologi MARA a for the grant of 600-IRMI/PERDANA 5/3 BESTARI (096/2018) as well as Faculty of Computer and Mathematical Sciences, Universiti Teknologi MARA, Shah Alam, Malaysia  ... 
doi:10.11591/ijai.v8.i3.pp264-269 fatcat:gvrdtsepkzbcnb5dubjqifhl3a

A Multimodal Anomaly Detector for Robot-Assisted Feeding Using an LSTM-based Variational Autoencoder [article]

Daehyung Park, Yuuna Hoshi, Charles C. Kemp
2017 arXiv   pre-print
The detection of anomalous executions is valuable for reducing potential hazards in assistive manipulation. Multimodal sensory signals can be helpful for detecting a wide range of anomalies.  ...  For evaluations with 1,555 robot-assisted feeding executions including 12 representative types of anomalies, our detector had a higher area under the receiver operating characteristic curve (AUC) of 0.8710  ...  However, the compressed representations of outliers (i.e., anomalous data) may be inliers in latent space.  ... 
arXiv:1711.00614v1 fatcat:lamgxvusrzh7xhg2s5npwatw4a

Combining machine learning and a universal acoustic feature-set yields efficient automated monitoring of ecosystems [article]

Sarab S. Sethi, Nick S. Jones, Ben D. Fulcher, Lorenzo Picinali, Dena J. Clink, Holger Klinck, C. David L. Orme, Peter H. Wrege, Robert M. Ewers
2019 biorxiv/medrxiv   pre-print
On the scale of seconds, we learned a typical soundscape model that allowed automatic identification of anomalous sounds in playback experiments, paving the way for real-time detection of irregular environmental  ...  Our highly generalisable approach, and the common set of features, will enable scientists to unlock previously hidden insights from eco-acoustic data and offers promise as a backbone technology for global  ...  Acknowledgements We would like to thank Till Hoffman for his input in selecting the audio features.  ... 
doi:10.1101/865980 fatcat:ukkix2iggbghxh2nvgi7ld4je4
« Previous Showing results 1 — 15 out of 13,925 results