Real Audio Research Articles

AN OVERVIEW AND ANALYSIS OF SILENT SOUND TECHNOLOGY

‌Nowadays whenever we are in a crowd or there was a to much background noise around us then it is very difficult to communicate or interact over the phone, but now we have an awesome solution on this problem that is ' silent sound technology ' it is a gift for those who lost their voice but wish to speak over the phone. this technology is produced by the Karlsruhe institute of technology, Germany and it will be available soon in the market. Silent sound technology is also known as ultrasonic sound or ultrasonic communication enables the transmission of data through inaudible sound waves. in this technology it detects every lip movement and convert it into the sound signal neglecting all background noise. without using real audio it is a visual interpretation of lip, mouth and face. So basically it reads our lip. This technology is available in English, German and French language, but not in language like Chinese because different tones hold different meanings in Chinese language. it use to allow silent calls and we can share our personal information like PIN , password or secret information to trusted one without knowing any other. it is also use in the Military for the secret missions. According to the researchers the technology have 99% efficiency only 1% possibility that it was produce a incorrect voice and upcoming 6-7 years the normal peoples are also use this technology because it launching very soon. Keywords : silent sound technology, electromyography, image processing.

AudioGest: Gesture-based Interaction for Virtual Reality using Audio Devices.

Current virtual reality (VR) system takes gesture interaction based on camera, handle and touch screen as one of the mainstream interaction methods, which can provide accurate gesture input for it. However, limited by application forms and the volume of devices, these methods cannot extend the interaction area to such surfaces as walls and tables. To address the above challenge, we propose AudioGest, a portable, plug-and-play system that detects the audio signal generated by finger tapping and sliding on the surface through a set of microphone devices without extensive calibration. First, an audio synthesis-recognition pipeline based on micro-contact dynamics simulation is constructed to generate modal audio synthesis from different materials and physical properties. Then the accuracy and effectiveness of the synthetic audio are verified by mixing the synthetic audio with real audio proportionally as the training sets. Finally, a series of desktop office applications are developed to demonstrate the application potential of AudioGest's scalability and versatility in VR scenarios.

Entry Count vs Occupancy Count to assess sector capacity with Fast Time Simulation

An air traffic controller should manage aircraft movements per hour to ensure the safety and efficiency of the traffic system. This study proposed an innovative methodology to estimate the air traffic controllers’ workload through sector occupancy. Starting with a model for the workload of sector controllers designed by Eurocontrol for tower controllers, the authors defined a taskload model from real audio that occurred in an examined Terminal Control Area (TMA). Moreover, trombones have been implemented to sequence arrival flows in the simulated TMA. Therefore, an airside model has been reproduced using Fast Time Simulation software, and its capacity has been calculated in terms of occupancy due to approaching movements. The results permit to evaluate the impact of infrastructural and operative modifications on the sector capacity in terms of hourly entry count and occupancy. Total and maximum occupancy values have been deepened to identify the best predictive factor of controllers’ workload and to manage the airspace configuration.

Wiener Filter and Deep Neural Networks: A Well-Balanced Pair for Speech Enhancement

This paper proposes a Deep Learning (DL) based Wiener filter estimator for speech enhancement in the framework of the classical spectral-domain speech estimator algorithm. According to the characteristics of the intermediate steps of the speech enhancement algorithm, i.e., the SNR estimation and the gain function, there is determined the best usage of the network at learning a robust instance of the Wiener filter estimator. Experiments show that the use of data-driven learning of the SNR estimator provides robustness to the statistical-based speech estimator algorithm and achieves performance on the state-of-the-art. Several objective quality metrics show the performance of the speech enhancement and beyond them, there are examples of noisy vs. enhanced speech available for listening to demonstrate in practice the skills of the method in simulated and real audio.

Dolphin vocal sound generation via deep WaveGAN

The marine biological sonar system evolved in the struggle of nature is far superior to the current artificial sonar. Therefore, the development of bionic underwater concealed detection is of great strategic significance to the military and economy. In this paper, a generative adversarial network (GAN) is trained based on the dolphin vocal sound dataset we constructed, which can achieve unsupervised generation of dolphin vocal sounds with global consistency. Through the analysis of the generated audio samples and the real audio samples in the time domain and the frequency domain, it can be proven that the generated audio samples are close to the real audio samples, which meets the requirements of bionic underwater concealed detection.

Audio Tampering Forensics Based on Representation Learning of ENF Phase Sequence

This paper proposes an audio tampering detection method based on the ENF phase and BI-LSTM network from the perspective of temporal feature representation learning. First, the ENF phase is obtained by discrete Fourier transform of ENF component in audio. Second, the ENF phase is divided into frames to obtain ENF phase sequence characterization, and each frame is represented as the change information of the ENF phase in a period. Then, the BI-LSTM neural network is used to train and output the state of each time step, and the difference information between real audio and tampered audio is obtained. Finally, these differences were fitted and dimensionally reduced by the fully connected network and classified by the Softmax classifier. Experimental results show that the performance of this method is better than the state-of-the-art approaches.

Sound Event Detection for Human Safety and Security in Noisy Environments

The objective of a sound event detector is to recognize anomalies in an audio clip and return their onset and offset. However, detecting sound events in noisy environments is a challenging task. This is due to the fact that in a real audio signal several sound sources co-exist and that the characteristics of polyphonic audio are different from isolated recordings. It is also necessary to consider the presence of noise (e.g. thermal and environmental). In this contribution, we present a sound anomaly detection system based on a fully convolutional network which exploits image spatial filtering and an Atrous Spatial Pyramid Pooling module. To cope with the lack of datasets specifically designed for sound event detection, a dataset for the specific application of noisy bus environments has been designed. The dataset has been obtained by mixing background audio files, recorded in a real environment, with anomalous events extracted from monophonic collections of labelled audios. The performances of the proposed system have been evaluated through segment-based metrics such as error rate, recall, and F1-Score. Moreover, robustness and precision have been evaluated through four different tests. The analysis of the results shows that the proposed sound event detector outperforms both state-of-the-art methods and general purpose deep learning-solutions.

Sound Event Detection by Pseudo-Labeling in Weakly Labeled Dataset.

Weakly labeled sound event detection (WSED) is an important task as it can facilitate the data collection efforts before constructing a strongly labeled sound event dataset. Recent high performance in deep learning-based WSED’s exploited using a segmentation mask for detecting the target feature map. However, achieving accurate detection performance was limited in real streaming audio due to the following reasons. First, the convolutional neural networks (CNN) employed in the segmentation mask extraction process do not appropriately highlight the importance of feature as the feature is extracted without pooling operations, and, concurrently, a small size kernel forces the receptive field small, making it difficult to learn various patterns. Second, as feature maps are obtained in an end-to-end fashion, the WSED model would be weak to unknown contents in the wild. These limitations would lead to generating undesired feature maps, such as noise in the unseen environment. This paper addresses these issues by constructing a more efficient model by employing a gated linear unit (GLU) and dilated convolution to improve the problems of de-emphasizing importance and lack of receptive field. In addition, this paper proposes pseudo-label-based learning for classifying target contents and unknown contents by adding ’noise label’ and ’noise loss’ so that unknown contents can be separated as much as possible through the noise label. The experiment is performed by mixing DCASE 2018 task1 acoustic scene data and task2 sound event data. The experimental results show that the proposed SED model achieves the best F1 performance with 59.7% at 0 SNR, 64.5% at 10 SNR, and 65.9% at 20 SNR. These results represent an improvement of 17.7%, 16.9%, and 16.5%, respectively, over the baseline.

Latent Vector Recovery of Audio GANs with Application in Deepfake Audio Detection

Advanced Generative Adversarial Networks (GANs) are remarkable in generating in- telligible audio from a random latent vector. In this paper, we examine the task of recovering the latent vector of both synthesized and real audio. We propose an auto- encoder inspired technique to train a deep residual neural network architecture to project audio synthesized by WaveGAN into the corresponding latent space with near identical reconstruction performance. To accommodate for the lack of an original latent vector for real audio, we optimize the residual network on a multi-level feature loss between the real audio samples and the reconstructed audio of the predicted latent vectors. In the case of synthesized audio, the Mean Squared Error (MSE) between the ground truth and recovered latent vector is minimized as well. We further investigated the audio reconstruction performance when several gradient optimization steps are applied to the predicted latent vector. Through our auto-encoder inspired method of training on real and synthesized audio, we are able to predict a latent vector that corresponds to a reasonable reconstruction of real audio. After training, we investigated the latent representations of real and synthesized audio files. Our analysis revealed distinct latent representational patterns for real and synthesized audio which can be used for deepfake audio detection. Even though we evaluated our method on WaveGAN, our proposed method is universal and can be applied to any other GANs.

A Swept-Sine-Type Single Measurement to Estimate Intermodulation Distortion in a Dynamic Range of Audio Signal Amplitudes

In the world of real audio systems, it is extremely important to model and identify their nonlinear behavior, especially in the case of professional audio devices. In this context, it is useful to have a quantitative estimation of the nonlinearity degree of the device, which can be obtained by exploiting an efficient and rapid measurement methodology. In this article, we propose an original estimation technique targeting the third-order intermodulation distortion (IMD) and based on a single detection. The proposed technique can be implemented both on devices operating in baseband and in bandpass. Starting from the same single detection, the technique allows to give either an estimate of the third-order IMD for the signal level actually used and to extrapolate the estimate of the IMD to signal levels different from the one actually used. Experimental verifications on real audio devices have allowed to validate the procedure in operational situations, thus confirming the validity of the proposed approach.

FABRICATION OF FUZZY LOGIC CONTROLLER ON SPEECH LABORATORY SYSTEM USING AT89C2051 MICROCONTROLLER

Background/Objectives: Most colleges and universities in the Philippines consider acquiring a speech laboratory equipment a tough matter to manage due to the prices and after-sales support. Methods/Statistical analysis: Many speech laboratory equipments from abroad exist but would not serve for a long time, particularly in terms of durability and maintainability. Findings: The researcher developed a digital innovation in developing speech laboratory equipment that suitable for the end-user to operate the system quickly using a personal computer (PC). An interface was connecting serial communication via USB port to a Fuzzy Logic Controller to facilitate the switching of speaker and microphone in the student cubicles/headsets in real audio while in peer-to-peer communication. Moreover, the monitor or console of the teacher operates in a more comfortable and user-friendly manner. When the teacher clicks the desired function or button, the Fuzzy Logic Controller module responds and processes the command to enable or disable the student's headset. On the student's side, there is a call button function to highlight the teacher console. Improvements/Applications: Hence, the system is maintainable and flexible that can even run on Windows Operating Systems and Linux.

Audio–visual domain adaptation using conditional semi-supervised Generative Adversarial Networks

Accessing large, manually annotated audio databases in an effort to create robust models for emotion recognition is a notably difficult task, handicapped by the annotation cost and label ambiguities. On the contrary, there are plenty of publicly available datasets for emotion recognition which are based on facial expressivity due to the prevailing role of computer vision in deep learning research, nowadays. Thereby, in the current work, we performed a study on cross-modal transfer knowledge between audio and facial modalities within the emotional context. More concretely, we investigated whether facial information from videos could be used to boost the awareness and the prediction tracking of emotions in audio signals. Our approach was based on a simple hypothesis: that the emotional state’s content of a person’s oral expression correlates with the corresponding facial expressions. Research in the domain of cognitive psychology was affirmative to our hypothesis and suggests that visual information related to emotions fused with the auditory signal is used from humans in a cross-modal integration schema to better understand emotions. In this regard, a method called dacssGAN (which stands for Domain Adaptation Conditional Semi-Supervised Generative Adversarial Networks) is introduced in this work, in an effort to bridge these two inherently different domains. Given as input the source domain (visual data) and some conditional information that is based on inductive conformal prediction, the proposed architecture generates data distributions that are as close as possible to the target domain (audio data). Through experimentation, it is shown that classification performance of an expanded dataset using real audio enhanced with generated samples produced using dacssGAN (50.29% and 48.65%) outperforms the one obtained merely using real audio samples (49.34% and 46.90%) for two publicly available audio–visual emotion datasets.

BER Computation of OFDM System Using Real Audio Signal for Rician Channel

Higher generation of mobile communication system required a very high bandwidth and it can be possible by using efficient modulation technique known as Orthogonal frequency division multiplexing (OFDM) .It has various applications in the field of wireless and wired application. Modeling of OFDM system has been done by using real data audio signal as an input for different fading channels. In this paper the modeling of OFDM technique is done using Rician fading channel .BER is computed for OFDM system using different values of SNR .The modeling of OFDM is done using MATLAB 8.1 simulink tool.

Spectral Analysis for Nonstationary Audio

A new approach for the analysis of nonstationary signals is proposed, with a focus on audio applications. Following earlier contributions, nonstationarity is modeled via stationarity-breaking operators acting on Gaussian stationary random signals. The focus is on time warping and amplitude modulation, and an approximate maximum-likelihood approach based on suitable approximations in the wavelet transform domain is developed. This paper provides theoretical analysis of the approximations, and introduces joint estimation of frequency, amplitude, and spectrum, a corresponding estimation algorithm. The latter is tested and validated on synthetic as well as real audio signal.

Alpha-Stable Matrix Factorization

Matrix factorization (MF) models have been widely used in data analysis. Even though they have been shown to be useful in many applications, classical MF models often fall short when the observed data are impulsive and contain outliers. In this study, we present $\alpha $ MF, a MF model with $\alpha $ -stable observations. Stable distributions are a family of heavy-tailed distributions that is particularly suited for such impulsive data. We develop a Markov Chain Monte Carlo method, namely a Gibbs sampler, for making inference in the model. We evaluate our model on both synthetic and real audio applications. Our experiments on speech enhancement show that $\alpha $ MF yields superior performance to a popular audio processing model in terms of objective measures. Furthermore, $\alpha $ MF provides a theoretically sound justification for recent empirical results obtained in audio processing.

Audio Classification based on Association and Hybrid Optimization Technique

pattern recognition areas and data mining, audio data classification is a most important topic. This paper describes a new classification method, where Optimal Classification Rule Extraction for multi-class Audio Data (O-cREAD). This classification method uses a new hybrid optimization approach for extracting optimal classification rules, and then these optimal rules are further used for classifying multi-class testing audio data to their respective classes with better accuracy. The optimal classification rule extraction is a two- step process. In the first step, frequent itemsets are generated by the hybrid apriori algorithm and generates classification rules using the association concept. Next, a new hybrid optimization approach is used for optimizing classification rules of classification method. The new hybrid optimization approach is based on Ant Colony Optimization (ACO) and Multi-Objective Genetic Algorithm (MOGA). The best feature of classification method (O-cREAD) is that size of classification rules can be dramatically reduced and produce more sophisticated or complicated rules to improve classification accuracy for classifying a real audio dataset into their respective classes. KeywordsData, Classification Rules, Hybrid Apriori Algorithm (H-AA), Genetic Algorithm (GA), Ant colony optimization (ACO), Hybrid optimization approach.

Soft Margin Based Low-Rank Audio Signal Classification

We propose an algorithm to do audio signal classification based on low-rank matrix representative audio data. Conventionally, the low-rank matrix data can be represented by a vector in high dimensional space. Some learning algorithms are then applied in such a vector space for matrix data classification. Particularly, maximum margin classifiers, such as support vector machine (SVM) etc. have received much attention due to their effectiveness. In this paper, we classify the data directly in the matrix space. Our methodology is built on recent studies about matrix classification with the trace norm constrained weight matrix and SVM's large-margin linear discrimination principle. The resulting low-rank SVM is then designed to maximize the margin between classes whilst minimizing the complexity of the classifier in both original and low-rank space. We compared our proposed algorithm with SVM and other state-of-the-art matrix signal classification methods. Experimental studies on real life audio signal classification show the effectiveness of our algorithm.

Practical design of delta-sigma multiple description audio coding

Abstract It was recently shown that delta-sigma quantization (DSQ) can be used for optimal multiple description (MD) coding of Gaussian sources. The DSQ scheme combined oversampling, prediction, and noise-shaping in order to trade off side distortion for central distortion in MD coding. It was shown that asymptotically in the dimensions of the resampling, prediction, and noise-shaping filters as well as asymptotically in the quantizer dimensions, all rate-distortion points on the symmetric quadratic Gaussian MD rate-distortion function could be achieved. In this work, we show that this somewhat theoretical framework is suitable for practical low-delay MD audio coding. In particular, we design a practical MD audio coder with two descriptions and provide simulations on real audio data. The simulations demonstrate that even when using low-dimensional noise-shaping, prediction, and resampling filters, it is possible to obtain good quality audio in the presence of packet losses. Simulations on real audio reveal that, contrary to existing designs, it is straightforward to obtain a large number of trade-off points between side distortion and central distortion, which makes the proposed coder suitable for a wide range of applications.

Particle filtering for TDOA based acoustic source tracking: Nonconcurrent Multiple Talkers

Room reverberation introduces multipath components into an audio signal and causes problems for acoustic source localization and tracking. Existing tracking methods based on the extended Kalman filter (EKF) and sequential importance resampling based particle filter (SIR-PF) usually assume that a single source is constantly active in the tracking scene. Assuming that multiple talkers may appear alternatively during a conversation, this paper develops an extended Kalman particle filtering (EKPF) approach for nonconcurrent multiple acoustic tracking (NMAT). Essentially, an EKF is introduced to obtain an optimum importance sampling, by which the particles are drawn according to the current time-delay of arrival (TDOA) measurements as well as the previous position estimates. Hence, the proposed approach can quickly adapt to the sharp position change when the source switches and the tracking lag in SIR-PF can be avoided. Moreover, the amplitude of the TDOA measurement is investigated to formulate a measurement hypothesis prior. Such a prior is fused into the tracking algorithm to enhance the tracking accuracy. Both simulations and real audio lab experiments are organized to study the tracking performance. The results demonstrate that the proposed EKPF approaches outperforms the SIR-PF and EKF in a broad range of tracking scenarios.

A Simple Prior for Audio Signals

We propose a simple prior for restoration problems involving oscillatory signals. The prior makes use of an underlying analytic frame decomposition with narrow subbands. Other than this, the prior does not have any other parameters, which makes it simple to use and apply. We demonstrate the utility of the proposed prior through some real audio restoration experiments.