Acoustic Event Detection System Research Articles

A Feature-Reduction Scheme Based on a Two-Sample t-Test to Eliminate Useless Spectrogram Frequency Bands in Acoustic Event Detection Systems

Acoustic event detection (AED) systems, combined with video surveillance systems, can enhance urban security and safety by automatically detecting incidents, supporting the smart city concept. AED systems mostly use mel spectrograms as a well-known effective acoustic feature. The spectrogram is a combination of frequency bands. A big challenge is that some of the spectrogram bands may be similar in different events and be useless in AED. Removing useless bands reduces the input feature dimension and is highly desirable. This article proposes a mathematical feature analysis method to identify and eliminate ineffective spectrogram bands and improve AED systems’ efficiency. The proposed approach uses a Student’s t-test to compare frequency bands of the spectrogram from different acoustic events. The similarity between each frequency band among events is calculated using a two-sample t-test, allowing the identification of distinct and similar frequency bands. Removing these bands accelerates the training speed of the used classifier by reducing the number of features, and also enhances the system’s accuracy and efficiency. Based on the obtained results, the proposed method reduces the spectrogram bands by 26.3%. The results showed an average difference of 7.77% in the Jaccard, 4.07% in the Dice, and 5.7% in the Hamming distance between selected bands using train and test datasets. These small values underscore the validity of the obtained results for the test dataset.

Deep Learning-based drone acoustic event detection system for microphone arrays

Deep Learning-based drone acoustic event detection system for microphone arrays

Impulse Acoustic Event Detection, Classification, and Localization System

To point out dangerous events is more and more necessary not only in military applications but also in civil areas. In hazardous situations, to detect, localize, and classify a source of danger (e.g., gunshot), an automatic acoustic measurement detection system may be preferable in favor of visual detection. This paper presents an automatic acoustic system that can detect, localize and classify acoustic impulse events, such as gunshots. The presented system is based on standalone units, which continuously monitor its surrounding. If an acoustic event is detected, an algorithm based on Mel Frequency Transformation and classification using a Support Vector Machine is performed. If the gunshot is recognized, the system calculates, using the Levenberg-Marquardt iterative solution algorithm, the exact location of the event and can estimate the caliber of the used firearm. The detection system is designed to monitor the surroundings around objects of interest, like campuses, hospitals, or public areas. The system has been tested by .22, 6.35 mm, 7,65 mm, and 9 mm short guns, and .22, 5.56 NATO, and 7.62 mm rifle guns with various subsonic and supersonic ammunition. The sets of diverse impulse acoustic events, like different slams, glass breaking, dog barking, or similar, have been used to train the classifier to avoid false alarms. The presented acoustic system can reliably detect, localize and classify hazardous acoustic events.

URBAN SOUND RECOGNITION USING DIFFERENT FEATURE EXTRACTION TECHNIQUES

The application of the advanced methods for noise analysis in the urban areas through the development of systems for classification of sound events significantly improves and simplifies the process of noise assessment. The main purpose of sound recognition and classification systems is to develop algorithms that can detect and classify sound events that occur in the chosen environment, giving an appropriate response to their users. In this research, a supervised system for recognition and classification of sound events has been established through the development of feature extraction techniques based on digital signal processing of the audio signals that are further used as an input parameter in the machine learning algorithms for classification of the sound events. Various audio parameters were extracted and processed in order to choose the best set of parameters that result in better recognition of the class to which the sounds belong. The created acoustic event detection and classification (AED/C) system could be further implemented in sound sensors for automatic control of environmental noise using the source classification that leads to reduced amount of required human validation of the sound level measurements since the target noise source is evidently defined.

Firearm Acoustic Detection in Hartford, Connecticut: Outcomes of a Trauma Center – Law Enforcement Collaboration

IntroductionFirearm homicide is a leading cause of violence-related death in the United States.Unfortunately, more than 80% of illegal firearm discharges are never reported to police by traditional means.ShotSpotterTM (Newark, California) is an acoustic firearm event detection system that can localize gunfire, prompting police, and subsequent emergency medical services (EMS) presence. Previously reported healthcare effects of acoustic detection are speculative in nature. We sought to investigate Hartford, Connecticut's experience with ShotSpotter​​​​​​​TM given its smaller size and broad coverage. MethodsThe three trauma centers in Hartford (two for adults and one for pediatric) collaborated with the Hartford Police to review outcomes of victims with acoustically detected gunshots and compare them to those who went undetected. We performed a retrospective review of patients who presented with gunshot wounds (GSW) over a 30-month period, from January 1, 2016 to June 30, 2018. Victim location and acoustic detection were reconciled by the police department and hospital staff independently. Patients were individually matched for location, prehospital response, treatment durations, and hospital outcomes.ResultsOf 387 GSW, 157 (40.6%) presented via EMS and were included in the sample. Of these, 89 correlated to a detection event (56.7%) and 68 had no correlating event (43.3%). These two groups had no difference in prehospital treatment times, scene and transport duration, and injury severity. Further, the need for surgery or transfusion, lengths of stay, and disposition, including mortality, did not differ.ConclusionsDespite limited previous reports demonstrating conferred benefits to acoustic detection of gunshots, Hartford’s experience showed no benefit. The potential for such systems to act as early warning systems is evident but may depend on a city’s resources, geography, and technology.

Ambient acoustic event assistive framework for identification, detection, and recognition of unknown acoustic events of a residence

In recent times, Ambient Assisted Living has emerged as Smart Living. Smart living is a subset of ambient intelligence, which uses the latest technologies, intellectual processes, and ambient intelligent methodologies to enable house residents to live independently with a virtual companion 24 × 7. Typically, these residents are highly engrossed in the daily routine activities that they tend to ignore certain acoustic events attributing them to the white noise caused due to tap water leakage, flush water leakage, the acoustics of door opening/closing, cupboard opening/closing, curtain opening/closing, television, shower, radio, chair and many more. These unattended events lead to a waste of critical energy resources such as electricity, water, and gas and may cause accidents in some cases. For the conducted experiments, a customized dataset termed as “unknown-2000” and ESC-50 has been used, which has more than 2000 audio sound classification samples. The customized dataset is used for the conducted experiments, consisting of various length acoustic events ranging from 2 s to 10 s. In the proposed review, we have identified, analyzed, and evaluated resident acoustic events using Librosa machine learning libraries, texture analysis using LBP methodology, LSTM-CNN, SVM, KNN, LSTM, Bi-LSTM, and Decision Tree-based classification approaches. Furthermore, in the proposed approach, based on the conducted rigorous and detailed analysis, we are also envisioning the prospective ways to enhance smart living concepts by proposing a novel Acoustic Event Detection and Classification System. The investigation results validate the success of the proposed approach. The obtained results indicate that the customized version of the LSTM-CNN based classification approach used in the conducted experiment has outperformed all the other customized classification approaches, such as SVM, KNN-based classification, C4.5 decision tree-based classification, LSTM, and Bi-LSTM based classification. The LSTM-CNN based classification model has achieved an average value of approximately 0.77 and a standard deviation of 0.2295. Furthermore, the obtained experiential results show that the proposed approach has produced a good performance in various noisy conditions such as SNR0, SNR3, SNR6, SNR9, SNR12, and SNR15. The system classification accuracy has been enhanced to 77% for various acoustic events of a residence. In the end, a detailed comparison of LBP and without LBP approaches has been carried out, which proves that the combination of LBP and LSTM-CNN classification approach provides better results than without the LBP classification approach. The proposed Ambient Acoustic Event Assistive Framework is a cost-effective alternative due to the use of low-cost microphone sensors in the conducted experiments.

Acoustic data augmentation for Mandarin-English code-switching speech recognition

Code-switching (CS) is a multilingual phenomenon where a speaker uses different languages in an utterance or between alternating utterances. Developing large-scale datasets for training code-switching acoustic and language models is challenging and extremely expensive. In this paper, we focus on the acoustic data augmentation for the Mandarin-English CS speech recognition task. Effectiveness of conventional acoustic data augmentation approaches are examined. More importantly, we propose a CS acoustic event detection system based on the deep neural network to extract real code-switching speech segments automatically. Then, the semi-supervised and active learning techniques are investigated to generate transcriptions of these segments. Finally, code-switching speech synthesis system is introduced to further enhance the acoustic modeling. Experimental results on the OC16-CE80 data, a Mandarin-English mixlingual speech corpus, demonstrate the effectiveness of the proposed methods.

Auxiliary Classifier Generative Adversarial Network With Soft Labels in Imbalanced Acoustic Event Detection

In acoustic event detection, the training data size of some acoustic events is often small and imbalanced. To deal with this, this paper proposes generating the virtual training data categorically using the auxiliary classifier generative adversarial networks. Soft labels of acoustic events are first calculated to represent the acoustic event localization information. The closer the current frame is to the middle of the manually labeled acoustic event, the higher the soft label will be, which makes the soft labels positively correlated with the acoustic event localization. Then, the acoustic event class and the quantized soft labels are used as the input condition to the auxiliary classifier generative adversarial networks to generate an arbitrary number of training samples. Experimental results on the TUT Sound Event 2016 under the home environment and TUT Sound Event 2017 under the street environment demonstrate the improved performance of the proposed technique compared to existing acoustic event detection systems.

A Survey: Neural Network-Based Deep Learning for Acoustic Event Detection

Recently, neural network-based deep learning methods have been popularly applied to computer vision, speech signal processing and other pattern recognition areas. Remarkable success has been demonstrated by using the deep learning approaches. The purpose of this article is to provide a comprehensive survey for the neural network-based deep learning approaches on acoustic event detection. Different deep learning-based acoustic event detection approaches are investigated with an emphasis on both strongly labeled and weakly labeled acoustic event detection systems. This paper also discusses how deep learning methods benefit the acoustic event detection task and the potential issues that need to be addressed for prospective real-world scenarios.

Random forest classification based acoustic event detection utilizing contextual-information and bottleneck features

The variety of event categories and event boundary information have resulted in limited success for acoustic event detection systems. To deal with this, we propose to utilize the long contextual information, low-dimensional discriminant global bottleneck features and category-specific bottleneck features. By concatenating several adjacent frames together, the use of contextual information makes it easier to cope with acoustic signals with long duration. Global and category-specific bottleneck features can extract the prior knowledge of the event category and boundary, which is ideally matched by the task of an event detection system. Evaluations on the UPC-TALP and ITC-IRST databases of highly variable acoustic events demonstrate the effectiveness of the proposed approaches by achieving a 5.30% and 4.44% absolute error rate improvement respectively compared to the state of art technique.

Pulse-coupled time synchronization for distributed acoustic event detection using wireless sensor networks

Time synchronization has proven to be critical in sensor fusion applications where the time of arrival is utilized as a decision variable. Herein, the application of pulse-coupled synchronization to an acoustic event detection system based on a wireless sensor network is presented. The aim of the system is to locate the source of acoustic events utilizing time of arrival measurements for different formations of the sensor network. A distributed localization algorithm is introduced that solves the problem locally using only a subset of the time of arrival measurements and then fuses the local guesses using averaging consensus techniques. It is shown that the pulse-coupled strategy provides the system with the proper level of synchronization needed to enable accurate localization, even when there exists drift between the internal clocks and the formation is not perfectly maintained. Moreover, the distributed nature of pulse-coupled synchronization allows coordinated synchronization and distributed localization over an infrastructure-free ad-hoc network.

Using multi-stream hierarchical deep neural network to extract deep audio feature for acoustic event detection

Extraction of effective audio features from acoustic events definitely influences the performance of Acoustic Event Detection (AED) system, especially in adverse audio situations. In this study, we propose a framework for extracting Deep Audio Feature (DAF) using multi-stream hierarchical Deep Neural Network (DNN). The DAF outputted from the proposed framework fuses the potential complementary information of multiple input feature streams and thus could be more discriminative than those input features for AED. We take two input feature streams and the hierarchical DNNs with two stages as an example for showing the extraction of DAF. The effectiveness of different audio features for AED is evaluated on two audio corpora, i.e. BBC (British Broadcasting Corporation) audio dataset and TV audio dataset with different signal-to-noise ratios. Experimental results show that DAF outperforms other features for AED under several experimental conditions.

Speech Technologies for Advanced Applications in Service Robotics

The multimodal interface for controlling functions of the complex modular robotic system, which can be deployed in difficult conditions as are rescue works, natural disasters, fires, decontamination purposes was designed. Such interface involves several fundamental technologies such as speech recognition, speech synthesis and dialogue management. To enable human operator to cooperate with designed robotic system, the sophisticated architecture was designed and described technologies were implemented. The automatic speech recognition system is introduced, which is based on Hidden Markov models and enables to control functions of the system using a set of voice commands. The text-to-speech system was prepared for producing feedback to the operator and dialogue manager technology was adopted, which makes it possible to perform the information exchange between operator and robotic system. The system proposed is enriched with acoustic event detection system, which consists of a set of five microphones integrated on the robotic vehicle, the post-processing unit and detection unit.

Acoustic event detection in meeting-room environments

Acoustic event detection (AED) aims at determining the identity of sounds and their temporal position in the signals that are captured by one or several microphones. The AED problem has been recently proposed for meeting-room or class-room environments, where a specific set of meaningful sounds has been defined, and several evaluations have been carried out (within the international CLEAR evaluation campaigns). This paper reports some work in AED done by the authors in that framework, and particularly presents the extension to the difficult problem of detecting overlapped sounds. Actually, temporal overlaps accounted for more than 70% of errors in the real-world interactive seminar recordings used in CLEAR 2007 evaluations. An attempt to deal with that problem at the level of models using our SVM-based AED system is reported in the paper. The proposed two-step system noticeably outperforms the baseline system for both an artificially generated database and a real seminar recording database. The databases and metrics developed for the CLEAR 2007 evaluations are also described. Finally, a real-time AED system implemented in the UPC’s smart-room using several microphones is reported, along with a GUI-based demo that includes also the output of an acoustic source localization system.