Minimally supervised sound event detection using a neural network

Detection and recognition of loud sounds and characteristic noises can significantly increase the level of safety and ensure timely response to various emergency situations. Audio event detection is the first step in recognizing audio signals in a continuous audio input stream. This article presents a number of problems that are associated with the development of sound event detection systems, such as the deviation for each environment and each sound category, overlapping audio events, unreliable training data, etc. Both methods for detecting monophonic impulsive audio event and polyphonic sound event detection methods which are used in the state-of-the-art sound event detection systems are presented. Such systems are presented in Detection and Classification of Acoustic Scenes and Events (DCASE) challenges and workshops, which take place every year. Beside a majority of works focusing on the improving overall performance in terms of accuracy many other aspects have also been studied. Several systems presented at DCASE 2021 task 4 were considered, and based on their analysis, there was a conclusion about possible future for sound event detection systems. Also the actual directions in the development of modern audio analytics systems are presented, including the study and use of various architectures of neural networks, the use of several data augmentation techniques, such as universal sound separation, etc.

Sound event detection in the forest environment is difficult to identify due to overlapping of noise and other sounds. Here, sound detection system in the forest needs more optimal sound recognition algorithms to provide accuracy in the results. The sound which is generating in the forest is contains more information in the signal and transmits in a different bands. Digital processing of sound signal is required to catch sound in a various pitches generating from variable distances. Different procedures and algorithms are available for extracting of feature of sound such as Liner Predictive Coding (LPC), Hidden Markov Model (HMM), Artificial Neural Network (ANN) and Mel Frequency Cepstral Coefficients (MFCCs). This paper presented the problems, issues and environmental constraints for sound event detection in the forest along with analysis on present algorithms.KeywordsLiner predictive codingMel frequency cepstral coefficientsHidden Markov model

| openaire: EC/H2020/637422/EU//EVERYSOUND

<p>Joint sound event localization and detection (SELD) is an emerging audio signal processing task adding spatial dimensions to acoustic scene analysis and sound event detection. A popular approach to modeling SELD jointly is using convolutional recur- rent neural network (CRNN) models, where CNNs learn high-level features from multi-channel audio input and the RNNs learn temporal relationships from these high-level features. However, RNNs have some drawbacks, such as a limited capability to model long temporal dependencies and slow training and inference times due to their sequential processing nature. Recently, a few SELD studies used multi-head self-attention (MHSA), among other innovations in their models. MHSA and the related transformer networks have shown state-of-the-art performance in various domains. While they can model long temporal dependencies, they can also be parallelized efficiently. In this paper, we study in detail the effect of MHSA on the SELD task. Specifically, we examined the effects of replacing the RNN blocks with self-attention layers. We studied the influence of stacking multiple self-attention blocks, using multiple attention heads in each self-attention block, and the effect of position embeddings and layer normalization. Evaluation on the DCASE 2021 SELD (task 3) development data set shows a significant improvement in all employed metrics compared to the baseline CRNN accompanying the task.</p>

Significance of analytic phase of speech signals in speaker verification

This work defines a new framework for performance evaluation of polyphonic sound event detection (SED) systems, which overcomes the limitations of the conventional collar-based event decisions, event F-scores and event error rates. The proposed framework introduces a definition of event detection that is more robust against labelling subjectivity. It also resorts to polyphonic receiver operating characteristic (ROC) curves to deliver more global insight into system performance than F1-scores, and proposes a reduction of these curves into a single polyphonic sound detection score (PSDS), which allows system comparison independently from operating points (OPs). The presented method also delivers better insight into data biases and classification stability across sound classes. Furthermore, it can be tuned to varying applications in order to match a variety of user experience requirements. The benefits of the proposed approach are demonstrated by re-evaluating the baseline and two of the top-performing systems from DCASE 2019 Task 4.

This paper presents a new technique for wheezes detection in respiratory sounds using a Long Short-Term Memory (LSTM), a specific type of Recurrent Neural Networks (RNNs). The purpose of this work is to develop an LSTM-based system and compare its classification performances to those obtained by a Multilayer Perceptron (MLP) feed-forward network. The MLP is a widely used neural network that has proven its efficiency in respiratory sound classification. Feed-forward networks do not consider time dependencies, while RNNs reach their limit in detecting dependencies when they occur at long time intervals. Because wheezing occurs over several consecutive intervals, we assume that LSTM takes into account the changing characteristics better than MLP and provides better results. Pulmonary sounds are characterized using the Mel-Frequency Cepstral Coefficients (MFCC) method before applying the LSTM-based classifier. As expected, the experimental tests show that LSTM takes advantage of the long-term dependencies observed in wheezing sounds to lead to better classification performances (accuracy of 91%).

The auditory brain circuits are biologically constructed to recognize and localize sounds by encoding a combination of cues that help individuals interpret sounds. The development of computational methods inspired by human capacities has established opportunities for improving machine hearing. Recent studies based on deep learning show that using convolutional recurrent neural networks (CRNNs) is a promising approach for sound event detection and localization in spatial sound. Nevertheless, depending on the sound environment, the performance of these systems is still far from reaching perfect metrics. Therefore, this work intends to boost the performance of state-of-the-art (SOTA) systems by using bio-inspired gammatone auditory filters and intensity vectors (IVs) for the acoustic feature extraction stage, along with the implementation of a temporal convolutional network (TCN) block into a CRNN model, to capture long term dependencies. Three data augmentation techniques are applied to increase the small number of samples in spatial audio datasets. The mentioned stages constitute our proposed Gammatone-based Sound Events Localization and Detection (G-SELD) system, which exceeded the SOTA results on four spatial audio datasets with different levels of acoustical complexity and with up to three sound sources overlapping in time.

Weakly labeled semi-supervised learning methods have recently drawn increasing attention from the research community for sound event detection tasks. Due to the weakness of the labelling, neural networks are often designed to perform sound event detection (SED) and audio tagging (AT) at the same time. In this paper, we propose a task-aware mean teacher method using a convolutional recurrent neural network (CRNN) with multi-branch structure to solve the SED and AT tasks differently. Specifically, a branch with coarse-level temporal resolution is designed for the AT task, while a branch with fine-level temporal resolution is designed for the SED task. The mean teacher based semi-supervised learning method is first adopted to improve the performance of the coarse-level AT branch by exploiting unlabeled data. Then the coarse-level AT branch is introduced as a teacher to guide the aggregated AT output of the fine-level SED branch, yielding an improvement in the SED performance. To further improve the AT and SED performance, information from multiple layers is exploited in the form of a multi-resolution feature. Experimental results on Task4 of the DCASE2018 challenge demonstrate the superiority of the proposed method, achieving 37.7% F1-score, which outperforms the winning system’s 32.4%.

A multilayer feedforward neural network with a modification to the standard backpropagation training of neural nets is considered. Information on the gradients of target outputs with respect to network inputs is used in this modification. This modified neural network is then applied for approximating design requirements of aerospace composite components, such as an aircraft engine guide vane and a satellite reflector assembly. This neural network functional approximator is then used with an optimization algorithm for determining the optimal parameters for component design requirements. The modified neural network provides for faster convergence of learning process and requires a smaller number of hidden layer nodes for convergence to similar training error levels. HE application of neural networks as universal approximators has been the focus of tremendous amount of activity in the past few years.16 Neural networks have been used for approximating and modeling both memoryless systems and dynamical systems with memory. This paper does not go into the various neural net architectures and training schemes, and the interested reader could refer to standard texts on this subject matter.1'2 This paper considers static neural networks (systems without memory) formed by interconnections without loops, called feedforward networks.2 In addition, this work considers a backpropagation learning scheme with adaptation of learning rate. There have been many additions to, or adaptations of, the basic backpropagation scheme, such as adaptive learning rate, learning with momentum, and the like,1'2 for improving convergence of the learning scheme, and these are not addressed in this paper. This paper explores a modification of the backpropagation learning scheme using information on the gradient of the target outputs with respect to the inputs of the network. The performance of the modified learning scheme is compared with that of the standard backpropagation learning scheme.2 A feedforward network with a single hidden layer trained with this modified learning scheme is applied in structural optimization problems. Figure 1 shows a feedforward net with a single hidden layer. The restriction to a single hidden layer is not a limitation of the modified learning scheme, and the new training scheme can be used on networks with multiple hidden layers also. The use of neural networks for structural response approximation has been reported by several authors.46 The primary reason for investigating such properties is to assess the applicability of trained neural networks for quick and accurate approximation of response functions during structural design optimization. Structural optimization, utilizing either gradient-based numerical search or genetic search algorithms, requires repeated function (response) evaluations during the search process. A single finite element analysis of large-scale structures requires a significant amount of computational time and resource and hence the need for robust approxima

In this paper, we propose a feature set by concatenating Mel-Frequency Cepstral Coefficients (MFCC) and Spectral Centroid Magnitude Coefficients (SCMC) features for Acoustic Scene Classification (ASC) using Deep Neural Networks (DNN). MFCC features are used to hold the acoustic characteristics such as spectral envelope of an acoustic scene in each frame. It also carries the sub-band average energy as a single dimension. SCMC features are used to hold the distribution of energy in a sub-band effectively. A test is carried out on Tampere University of Technology (TUT) Acoustic Scenes 2017 Dataset. The DNN architecture for utterance level classification has been used. The proposed system’s performance on a 4-fold cross-validation setup is 80.2% and it gives 5.4% relative improvement in performance when compared to the baseline system that uses log-Mel band energies with Multi-Layer Perceptron model.

This paper presents and discusses various metrics proposed for evaluation of polyphonic sound event detection systems used in realistic situations where there are typically multiple sound sources active simultaneously. The system output in this case contains overlapping events, marked as multiple sounds detected as being active at the same time. The polyphonic system output requires a suitable procedure for evaluation against a reference. Metrics from neighboring fields such as speech recognition and speaker diarization can be used, but they need to be partially redefined to deal with the overlapping events. We present a review of the most common metrics in the field and the way they are adapted and interpreted in the polyphonic case. We discuss segment-based and event-based definitions of each metric and explain the consequences of instance-based and class-based averaging using a case study. In parallel, we provide a toolbox containing implementations of presented metrics.

Speaker identification refers to the process of recognizing human voice using artificial intelligence techniques. Speaker identification technologies are widely applied in voice authentication, security and surveillance, electronic voice eavesdropping, and identity verification. In the speaker identification process, extracting discriminative and salient features from speaker utterances is an important task to accurately identify speakers. Various features for speaker identification have been recently proposed by researchers. Most studies on speaker identification have utilized short-time features, such as perceptual linear predictive (PLP) coefficients and Mel frequency cepstral coefficients (MFCC), due to their capability to capture the repetitive nature and efficiency of signals. Various studies have shown the effectiveness of MFCC features in correctly identifying speakers. However, the performances of these features degrade on complex speech datasets, and therefore, these features fail to accurately identify speaker characteristics. To address this problem, this study proposes a novel fusion of MFCC and time-based features (MFCCT), which combines the effectiveness of MFCC and time-domain features to improve the accuracy of text-independent speaker identification (SI) systems. The extracted MFCCT features were fed as input to a deep neural network (DNN) to construct the speaker identification model. Results showed that the proposed MFCCT features coupled with DNN outperformed existing baseline MFCC and time-domain features on the LibriSpeech dataset. In addition, DNN obtained better classification results compared with five machine learning algorithms that were recently utilized in speaker recognition. Moreover, this study evaluated the effectiveness of one-level and two-level classification methods for speaker identification. The experimental results showed that two-level classification presented better results than one-level classification. The proposed features and classification model for identifying a speaker can be widely applied to different types of speaker datasets.

In the field of medicine, with the introduction of computer systems that can collect and analyze massive amounts of data, many non-invasive diagnostic methods are being developed for a variety of conditions. In this study, our aim is to develop a non-invasive method of classifying respiratory sounds that are recorded by an electronic stethoscope and the audio recording software that uses various machine learning algorithms.In order to store respiratory sounds on a computer, we developed a cost-effective and easy-to-use electronic stethoscope that can be used with any device. Using this device, we recorded 17,930 lung sounds from 1630 subjects.We employed two types of machine learning algorithms; mel frequency cepstral coefficient (MFCC) features in a support vector machine (SVM) and spectrogram images in the convolutional neural network (CNN). Since using MFCC features with a SVM algorithm is a generally accepted classification method for audio, we utilized its results to benchmark the CNN algorithm. We prepared four data sets for each CNN and SVM algorithm to classify respiratory audio: (1) healthy versus pathological classification; (2) rale, rhonchus, and normal sound classification; (3) singular respiratory sound type classification; and (4) audio type classification with all sound types. Accuracy results of the experiments were; (1) CNN 86%, SVM 86%, (2) CNN 76%, SVM 75%, (3) CNN 80%, SVM 80%, and (4) CNN 62%, SVM 62%, respectively.As a result, we found out that spectrogram image classification with CNN algorithm works as well as the SVM algorithm, and given the large amount of data, CNN and SVM machine learning algorithms can accurately classify and pre-diagnose respiratory audio.

Feature extraction techniques for input representation to diagnose infant diseases have received significant attention recently. Mel Frequency Cepstral Coefficients (MFCC) is one of the most popular feature extraction techniques due to its representation method being very similar to the human auditory system. The MFCC method for feature extraction depends on several important parameter settings, namely the number of filter banks, and the number of coefficients used in the final representation. These settings affects the way the features are represented, and in turn, affects the performance of the classifier for diagnosis of the disease. In this paper, the Particle Swarm Optimization (PSO) algorithm was used to optimise the parameters of the MFCC feature extraction method for classifying infants with asphyxia. The extracted MFCC features were then used to train several MLP classifiers over different initialization values. The accuracy of these classifiers was then used to guide the PSO optimization. Our results show that the optimization of MFCC computation using PSO yielded 93.9% accuracy, an improvement of 1.45% over typical MFCC parameter settings using the same classifier.