Audio Applications Research Articles

Comparative evaluation of omnidirectional and directional micro-electromechanical system microphone performance

As the need for directionality becomes a key requirement in audio applications, directional microphones have begun to enter the micro-electromechanical system (MEMS) design and market space, and their performance is approaching that of top-of-the-line omnidirectional MEMS microphones. This presentation examines and compares the performance limitations for both types of MEMS microphones and suggests more comprehensive methods of characterization that allow the qualities of directional MEMS microphones to be fully captured. Mechanical thermal noise caused by Brownian motion of air particles, measured with a laser Doppler vibrometer system, are shared and discussed for a variety of microphones and test structures. Parameters such as signal-to-noise ratio, total harmonic distortion, and frequency response are re-examined in order to capture the value of a microphone that exhibits directional behavior. Insights and suggestions are made for improvement and future work.

Implementing neural networks in low-latency audio applications

The use of neural networks is becoming increasingly prevalent due to their ability to represent complex relationships and solve complex problems. However, implementing these models in systems that require low-latency output can be challenging, especially for practitioners who are used to developing their models in controlled environments like Python notebooks. Another issue is the high computational cost of complex models, which limits the minimum possible latency. This paper presents approaches for deploying models in audio applications, discusses the advantages and disadvantages of each approach, and presents strategies to reduce the inference cost of models without significantly sacrificing accuracy, using techniques such as model quantization. To illustrate these methods, example implementations of real-time beamforming deconvolution and real-time music DSP processing are shown.

Spatial sigma-delta modulation for coarsely quantized linear, circular, and spherical arrays

Quantization noise shaping, also known as sigma-delta modulation, is widely used in analog-to-digital converters. Such converters oversample a signal in time, quantize it to one or a few bits of precision, and then use analog feedback to shape quantization noise outside the band of interest, producing a high-precision output. Similarly, if a sensor array is spatially oversampled so that its elements are much less than one-half wavelength apart, then quantization noise can be propagated between channels to shape its spatial distribution. A beamformer acts as a spatial filter, removing shaped quantization noise to produce a high-precision output even with coarsely quantized sensors. Spatial sigma-delta modulation has been explored primarily for radio-frequency applications using linear arrays. In this presentation, we apply spatial noise shaping to dense acoustic sensor arrays, including circular and spherical arrays. For immersive audio applications, we demonstrate how noise-shaping feedback can be used to push quantization noise into higher-order Ambisonic modes.

A compact high performance micro-electro-mechanical systems microphone

This paper reports on the development of a MEMS capacitive microphone design with 72 dBA signal-to-noise-ratio (SNR) in a compact 3.4 × 2.3 × 0.7 mm3package. The design incorporates a circular diaphragm disc suspended on one end of the cantilever beam. The diaphragm, under the bias conditions, is supported using peripheral and center protrusions extended from the back plate. The design optimization is targeted to achieve high sensitivity and low damping noise to achieve maximum SNR possible in the mentioned footprint. Finite element modeling (FEM) combined with the lumped element circuit modeling have been implemented to realize the microphone performance. The results have been validated against the measurement with very good correlation of sensitivity, noise and total harmonic distortion (THD). With the sensitivity of −35 dBV (ref. 1 V/Pa at 1 kHz) and acoustic overload point of 134 dBSPL, this is one of the highest performing MEMS analog microphone reported today. Therefore, it is very well suited for audio applications such as mobile phones, true wireless stereo (TWS) earphones, hearing aids and automotive, which demand miniaturized size, low cost and high performance.

Fast and accurate modeling of parametric array loudspeakers in the frequency domain

The parametric array loudspeaker (PAL) is known for its ability to generate highly directional audio beams. It has been widely used in a variety of audio applications, such as the sound field reproduction and active noise control. Fast and accurate computations of the audio sound field generated by a PAL are challenging due to the complexity of nonlinear interactions between airborne ultrasonic waves. This paper presents an overview of the commonly used computational methods in the frequency domain under the quasilinear assumption, which include the direct integration method, convolution directivity model, Gaussian beam expansions, cylindrical expansions, spherical expansions, and finite element methods. The computational efficiency and the accuracy of these methods in the near field, Westervelt far field, and the inverse-law far field are compared. Future challenges and perspectives are discussed.

MiniatureVQNet: A Light-Weight Deep Neural Network for Non-Intrusive Evaluation of VoIP Speech Quality

In IP audio systems, audio quality is degraded by environmental noise, poor network quality, and encoding–decoding algorithms. Therefore, there is a need for a continuous automatic quality evaluation of the transmitted audio. Speech quality monitoring in VoIP systems enables autonomous system adaptation. Furthermore, there are diverse IP audio transmitters and receivers, from high-performance computers and mobile phones to low-memory and low-computing-capacity embedded systems. This paper proposes MiniatureVQNet, a single-ended speech quality evaluation method for VoIP audio applications based on a lightweight deep neural network (DNN) model. The proposed model can predict the audio quality independent of the source of degradation, whether noise or network, and is light enough to run in embedded systems. Two variations of the proposed MiniatureVQNet model were evaluated: a MiniatureVQNet model trained on a dataset that contains environmental noise only, referred to as MiniatureVQNet–Noise, and a second model trained on both noise and network distortions, referred to as MiniatureVQNet–Noise–Network. The proposed MiniatureVQNet model outperforms the traditional P.563 method in terms of accuracy on all tested network conditions and environmental noise parameters. The mean squared error (MSE) of the models compared to the PESQ score for ITU-T P.563, MiniatureVQNet-Noise, and MiniatureVQNet–Noise–Network was 2.19, 0.34, and 0.21, respectively. The performance of both the MiniatureVQNet–Noise–Network and MiniatureVQNet-Noise model depends on the noise type for an SNR greater than 0 dB and less than 10 dB. In addition, training on a noise–network-distorted speech dataset improves the model prediction accuracy in all VoIP environment distortions compared to training the model on a noise-only dataset.

Setup of Consumer Wearable Devices for Exposure and Health Monitoring in Population Studies

Wearable sensors, which are often embedded in commercial smartwatches, allow for continuous and non-invasive health measurements and exposure assessments in clinical studies. Nevertheless, the real-life application of these technologies in studies involving a large number of participants for a significant observation period may be hindered by several practical challenges. In this study, we present a modified protocol from a previous intervention study for the mitigation of health effects from desert dust storms. The study involved two distinct population groups: asthmatic children aged 6-11 years and elderly patients with atrial fibrillation (AF). Both groups were equipped with a smartwatch for the assessment of physical activity (using a heart rate monitor, pedometer, and accelerometer) and location (using GPS signals to locate individuals in indoor "at home" or outdoor microenvironments). The participants were required to wear the smartwatch equipped with a data collection application on a daily basis, and data were transmitted via a wireless network to a centrally administered data collection platform for the near real-time assessment of compliance. Over a period of 26 months, more than 250 children and 50 patients with AF participated in the aforementioned study. The main technical challenges identified included restricting access to standard smartwatch features, such as gaming, internet browser, camera, and audio recording applications, technical issues, such as loss of GPS signal, especially in indoor environments, and the internal smartwatch settings interfering with the data collection application. The aim of this protocol is to demonstrate how the use of publicly available application lockers and device automation applications made it possible to address most of these challenges in a simple and cost-effective way. In addition, the inclusion of a Wi-Fi received signal strength indicator significantly improved indoor localization and largely minimized GPS signal misclassification. The implementation of these protocols during the roll-out of this intervention study in the spring of 2020 led to significantly improved results in terms of data completeness and data quality.

Setup of Consumer Wearable Devices for Exposure and Health Monitoring in Population Studies.

Wearable sensors, which are often embedded in commercial smartwatches, allow for continuous and non-invasive health measurements and exposure assessments in clinical studies. Nevertheless, the real-life application of these technologies in studies involving a large number of participants for a significant observation period may be hindered by several practical challenges. In this study, we present a modified protocol from a previous intervention study for the mitigation of health effects from desert dust storms. The study involved two distinct population groups: asthmatic children aged 6-11 years and elderly patients with atrial fibrillation (AF). Both groups were equipped with a smartwatch for the assessment of physical activity (using a heart rate monitor, pedometer, and accelerometer) and location (using GPS signals to locate individuals in indoor "at home" or outdoor microenvironments). The participants were required to wear the smartwatch equipped with a data collection application on a daily basis, and data were transmitted via a wireless network to a centrally administered data collection platform for the near real-time assessment of compliance. Over a period of 26 months, more than 250 children and 50 patients with AF participated in the aforementioned study. The main technical challenges identified included restricting access to standard smartwatch features, such as gaming, internet browser, camera, and audio recording applications, technical issues, such as loss of GPS signal, especially in indoor environments, and the internal smartwatch settings interfering with the data collection application. The aim of this protocol is to demonstrate how the use of publicly available application lockers and device automation applications made it possible to address most of these challenges in a simple and cost-effective way. In addition, the inclusion of a Wi-Fi received signal strength indicator significantly improved indoor localization and largely minimized GPS signal misclassification. The implementation of these protocols during the roll-out of this intervention study in the spring of 2020 led to significantly improved results in terms of data completeness and data quality.

LOLA SDR: Low Power Low Latency Software Defined Radio for Broadcast Audio Applications

This brief proposes a new low power, low latency and low cost reconfigurable architecture for software defined radio. Due to their flexibility and reconfigurability, software defined radios are now massively used as wideband transceivers, channel sounders or network gateways. However, they often struggle to meet the desired requirements in terms of energy consumption and throughput. In this brief, we present a new architecture capable of tackling these challenges, by combining an off-the-shelf generic radio component with a low power microcontroller associated to a Fourier transform coprocessor. To prove the benefit of our approach, after describing the key assets of the architecture, we derive a complete physical layer dedicated to audio broadcast applications. This chain is capable of streaming High Definition audio stream in real time with low power (437mW) and very low latency (854 <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\mu \text{s}$ </tex-math></inline-formula> ). We show that our processing chain can be flawlessly run on our architecture paving the way for larger adoption of a new generation of low power low latency software defined radio architectures.

Design of causal preconditioning filters for adaptive filtering algorithms in real-time multi-channel active noise control applications

Least-mean-square (LMS) algorithm has been widely used in the area of active noise control (ANC). One practical concern of LMS-based algorithms is its slow convergence rate in multi-channel broadband noise control applications. To improve the convergence speed of traditional LMS algorithm, preconditioning filters were added to the LMS system in previous studies to remove the correlation between the reference signals and decouple the plant responses. However, the preconditioning filters implemented previously are usually obtained through singular value decomposition of the cross spectral matrix of reference signals and the plant response matrix, which does not lead to causal preconditioning filters and, thus, it can only be applied in spatial audio applications when delay is not an important concern, but cannot be implemented in real-time active noise control applications. In the current work, a method is proposed to obtain a casual multi-channel preconditioning filter for real-time active noise control applications through a numerically robust algorithm to perform a spectral factor decomposition to the reference signal cross spectral matrix and a minimum-phase and all-pass decomposition to the multi-channel secondary path. Simulation results show that, by applying the proposed preconditioning filter, the convergence speed of the LMS algorithm can be significantly improved.

Ideological and Political Course Teaching in Colleges and Universities Based on the Construction of Computer Multimedia Technology

This system is under the development mode of the WeChat public platform, using the Unified Modeling Language (UML) as the description language for analysis and design, combined with the mature library-level basic platform J2EE and the lightweight open source database Mysql design and implementation of a B/ S-structured library book personalized recommendation system. Learning the construction mode of audio reading application software or audio reading WeChat applet in the market, starting from the product construction principles, supporting elements, page design scheme and operation scheme, etc., put forward a comprehensive method for university libraries to use applet to build an audio reading service platform sexual solutions.

A sound source localization method based on improved second correlation time delay estimation

The sound source localization (SSL) system based on the microphone array has important applications in audio and video conference, security monitoring and intelligent cockpit. However, the SSL method based on time difference of arrival is susceptible to ambient noise. Therefore, an improved second correlation delay estimation algorithm is proposed in this paper. The pure source signal is obtained by wavelet denoising, and then the time delay is calculated by the second correlation time delay estimation algorithm with the weighting functions of the smoothed coherence transform and the Roth processor. The position of the sound target is calculated from the time delay. Aiming at the moving target, an extended Kalman filter is introduced to track the moving trajectory of the sound source. The static and moving SSL simulations are conducted and the results of the proposed algorithm are compared with those of the single-weighted quadratic correlation (SQC) algorithm and the high-power quadratic correlation algorithm. The static sound source positioning errors of the proposed algorithm under −10 dB SNR are respectively 3.97 m and 5.86 m smaller than those of the HQC algorithm and the SQC algorithm. The moving SSL trajectory based on the proposed algorithm is still closest to the real track under −10 dB SNR. This indicates that the proposed algorithm has high precision and strong robustness for sound source location in the low signal-to-noise ratio (SNR) environment. In the experiment, the proposed algorithm can accurately calculate the direction of arrival (DOA) of static sound source and stably track DOA of moving sound source. This is consistent with the simulation results, which further verifies the effectiveness and practicability of the algorithm. This novel algorithm with high time delay estimation accuracy is of great significance for SSL in low SNR environment.

O USO DE PODCASTS COMO INSTRUMENTO DIDÁTICO NA EDUCAÇÃO: ABORDAGENS NOS PERIÓDICOS NACIONAIS ENTRE 2009 E 2020

ABSTRACT: The current stage of technology development achieved by society allows us to state that we have been living in a fourth or even fifth industrial revolution. In that context, it is necessary to consider how education, and mostly school, as one of the places of teaching and learning, has been adapted to that new ecosystem. One of the tools that emerges from this context is the podcast, a tool that allows sharing audio in applications and hosting sites. In that way, the purpose of this work was to evaluate how podcasts have emerged and expanded in educational environments in Brazil, mapping the national production of articles through the Capes (Centro de Aperfeiçoamento de Pessoal de Nível Superior) journals portal. At last, eighteen articles published between 2009 and 2020 were selected in order to form the corpus of analysis, then the institution of affiliation of the authors, their graduation areas, type of publication, most cited journals, and the chronological distribution of these citations were verified. Such data allowed us to establish that the attempts to use podcasts in educational environments have been increasing significantly between 2013 and 2020. Regarding the educational institutions, public ones stood out to the detriment of the private ones, it is also important to emphasize that 40,4% of the citations made by the authors of the corpus are books, while 28.9% are journals, indicating that this is still an area with a lot of space for research, since most books citations refer to single authors.

Meta-AF: Meta-Learning for Adaptive Filters

Adaptive filtering algorithms are pervasive throughout signal processing and have had a material impact on a wide variety of domains including audio processing, telecommunications, biomedical sensing, astrophysics and cosmology, seismology, and many more. Adaptive filters typically operate via specialized online, iterative optimization methods such as least-mean squares or recursive least squares and aim to process signals in unknown or nonstationary environments. Such algorithms, however, can be slow and laborious to develop, require domain expertise to create, and necessitate mathematical insight for improvement. In this work, we seek to improve upon hand-derived adaptive filter algorithms and present a comprehensive framework for learning online, adaptive signal processing algorithms or update rules directly from data. To do so, we frame the development of adaptive filters as a meta-learning problem in the context of deep learning and use a form of self-supervision to learn online iterative update rules for adaptive filters. To demonstrate our approach, we focus on audio applications and systematically develop meta-learned adaptive filters for five canonical audio problems including system identification, acoustic echo cancellation, blind equalization, multi-channel dereverberation, and beamforming.We compare our approach against common baselines and/or recent state-of-the-art methods. We show we can learn high-performing adaptive filters that operate in real-time and, in most cases, significantly outperform each method we compare against – all using a single general-purpose configuration of our approach.

A Universal Audio Steganalysis Scheme Based on Multiscale Spectrograms and DeepResNet

Given the popularity of audio and video applications, compressed audio has become an important carrier of covert communication on the Internet. Many novel compressed audio steganography schemes have emerged that offer good hiding capability and aural concealment. In this paper, a universal steganalysis scheme called MultiSpecNet is proposed to detect steganography based on multiple embedding domains (advanced audio coding (AAC) and MPEG-1 Audio Layer III (MP3)), which are currently the two most popular compressed audio standards. The basic idea is that modification of either domain by a steganography scheme will change the time-frequency relationship of the audio signal after decoding. The proposed approach adopts the spectrogram as the input feature to extract richer information. DeepResNet is used to learn the distinguishing feature representations, and multiscale spectrograms are used to enrich the feature diversity. The experimental results show that the proposed scheme is effective at detecting different steganography schemes based on the AAC and MP3 embedding domains. The detection accuracy of the proposed scheme is higher than that achieved by other state-of-the-art schemes. Using spectrograms as the input, DeepResNet achieves better performance than schemes using quantized modified discrete cosine transform (MDCT) coefficients and mel-spectrogram, although the quantized MDCT coefficient is the parameter modified by the steganography schemes directly and mel-spectrogram is very popular and effective for general audio signal analysis. To the best of our knowledge, this work is the first audio steganalysis scheme that can detect multiple steganography schemes in both the MP3 and AAC embedding domains. The method proposed in this paper can be extended to audio steganalysis for other codecs or for audio forensics purposes.

Continuous-Time Analog Filters for Audio Edge Intelligence: Review on Circuit Designs [Feature

Edge audio devices can reduce data bandwidth requirements by pre-processing input speech on the device before transmission to the cloud. As edge devices are required to ensure always-on operation, their stringent power constraints pose several design challenges and force IC designers to look for solutions that use low standby power. One promising bio-inspired approach is to combine the continuous-time analog filter channels with a small memory footprint deep neural network that is trained on edge tasks such as keyword spotting, thereby allowing all blocks to be embedded in an IC. This paper reviews the historical background of the continuous-time analog filter circuits that have been used as feature extractors for current edge audio devices. Starting from the interpretation of a basic biquad filter as a two-integrator-loop topology, we introduce the progression in the design of second-order low-pass and band-pass filters ranging from OTA-based to source-follower-based architectures. We also derive and analyze the small-signal transfer function and discuss their usage in edge audio applications.

Wave Digital Models of Piezoelectric Transducers for Audio Applications

The number of consumer electronics devices integrating piezoelectric (PE) transducers as flat-panel loudspeakers has recently experienced a steep increase. Being very thin, in fact, PE transducers well cope with the miniaturization process that characterizes the market. However, at the same time, their reduced dimensions cause the sound pressure level to be poor at low frequencies, impairing the overall acoustic response. In this article, we derive novel efficient discrete-time models of PE transducers in the wave digital (WD) domain such that they can be integrated in the future into signal processing algorithms for audio output enhancement. WD filters are, in fact, making headway among audio digital signal processing techniques based on circuit equivalent models thanks to their efficiency, robustness, and accuracy. Starting from circuital representations of the piezo constitutive equations, we derive both lumped and distributed models. In particular, we show how it is possible to implement the frequency-dependent elements characterizing Mason’s model in the WD domain, contrary to what can be done using mainstream SPICE-like simulators. Such WD implementations come in handy for the simulation and fast prototyping of PE systems, paving the way toward the design of model-based digital signal processing algorithms for enhancing the transducer acoustic performance.

Low-complexity energy-aware sensor selection for noise reduction in distributed microphone networks

Noise reduction (NR) is a necessary front-end in many audio applications for improving signal quality. It was shown that sparsity-promoting sensor selection potentially makes a trade-off between energy consumption and NR performance, which is rather important for large-scale wireless acoustic sensor networks (WASNs), where many sensors contribute negligibly to NR but energy consumption affects the lifetime of WASNs. This paper presents a sensor selection approach for beamforming-based NR by minimizing the total energy consumption and constraining the output noise variance. Motivated by the optimal semi-definite programming (SDP) solution and the utility-based method, we propose three low-complexity selection metrics: weighted utility, gradient, and weighted input signal-to-noise ratio (SNR). It is shown that the proposed weighted utility and gradient-based methods are near-optimal in performance but much faster than the SDP-based method, and the weighted SNR method has the lowest time complexity with a tiny performance sacrifice. Numerical results using a simulated WASN validate the superiority of the proposed approaches over conventional methods.

Study on Low-jitter and Low-power PLL Architectures for Mobile Audio Systems

This paper compares four phase-locked loops (PLLs) for mobile audio applications. We compare and analyze PLL structures and discuss the optimized PLL structure in the audio band frequency. A charge-pump-based integer-N PLL (NPLL) is employed as a reference. To improve the jitter performances, multiplying delay-locked loop (MDLL), sub-sampling PLL (SSPLL), and reference-sampling PLL (RSPLL) are employed and analyzed. The frequency range of the PLLs is from 8 MHz to 71.5 MHz. These PLL chips are fabricated using a Samsung 0.13-μm CMOS process. The resulting figures-of-merit for the NPLL, MDLL, SSPLL, and RSPLL are -204.3, -211.07, -220.29, and -213.32 dB, respectively, at 24.576 MHz. The total power consumption from a 1.5-V supply voltage is 1.82, 1.35, 1.43, and 1.64 mW, respectively.

Dual quaternion ambisonics array for six-degree-of-freedom acoustic representation

Spatial audio methods are gaining a growing interest due to the spread of immersive audio experiences and applications, such as virtual and augmented reality. For these purposes, 3D audio signals are often acquired through arrays of Ambisonics microphones, each comprising four capsules that decompose the sound field in spherical harmonics. In this paper, we propose a dual quaternion representation of the spatial sound field acquired through an array of two First Order Ambisonics (FOA) microphones. The audio signals are encapsulated in a dual quaternion that leverages quaternion algebra properties to exploit correlations among them. This augmented representation with 6 degrees of freedom (6DOF) involves a more accurate coverage of the sound field, resulting in a more precise sound localization and a more immersive audio experience. We evaluate our approach on a sound event localization and detection (SELD) benchmark. We show that our dual quaternion SELD model with temporal convolution blocks (DualQSELD-TCN) achieves better results with respect to real and quaternion-valued baselines thanks to our augmented representation of the sound field. Full code is available at: https://github.com/ispamm/DualQSELD-TCN.