The 50 most-cited articles in The Journal of the Acoustical Society of America, 2021

Metamaterials are attracting increasing interest in the field of acoustics due to their sound insulation effects. By periodically arranged structures, acoustic metamaterials can influence the way sound propagates in acoustic media. To date, the design of acoustic metamaterials relies primarily on the expertise of specialists since most effects are based on localized solutions and interference. This paper outlines a deep learning-based approach to extend current knowledge of metamaterial design in acoustics. We develop a design method by using conditional generative adversarial networks. The generative network proposes a cell candidate regarding a desired transmission behavior of the metamaterial. To validate our method, numerical simulations with the finite element method are performed. Our study reveals considerable insight into design strategies for sound insulation tasks. By providing design directives for acoustic metamaterials, cell candidates can be inspected and tailored to achieve desirable transmission characteristics.

Distributed acoustic sensing (DAS) is a recent instrumental approach allowing the conversion of fiber-optic cables into dense arrays of acoustic sensors. This technology is attractive in marine environments where instrumentation is difficult to implement. A promising application is the monitoring of environmental and anthropic noise, leveraging existing telecommunication cables on the seafloor. We assess the ability of DAS to monitor such noise using a 41.5 km-long cable offshore of Toulon, France, focusing on a known and localized source. We analyze the noise emitted by the same tanker cruising above the cable, first 5.8 km offshore in 85 m deep bathymetry, and then 20 km offshore, where the seafloor is at a depth of 2000 m. The spectral analysis, the Doppler shift, and the apparent velocity of the acoustic waves striking the fiber allow us to separate the ship radiated noise from other noise. At 85 m water depth, the signal-to-noise ratio is high, and the trajectory of the boat is recovered with beamforming analysis. At 2000 m water depth, although the acoustic signal of the ship…

The noise attenuation properties of an acoustic liner consisting of Helmholtz resonators with extended necks (HRENs) are investigated. An optimal liner constructed by 16 inhomogeneous HRENs is designed to be effective in sound absorption in a prescribed frequency range from 700 to 1000 Hz. Its quasi-perfect absorption capability (average absorption coefficient above 0.9) is validated by measurements and simulations. The resonance frequencies of the individual resonators in the designed liner are just located within the effective absorption bandwidth, indicating the overlapping phenomenon of absorption peaks. In addition, the liner maintains a thin thickness, about 1/25th with respect to the longest operating wavelengths. To assess the acoustic performance of the designed liner in the presence of mean flow, experimental investigations are performed in a flow tube. Results show a near flat transmission loss is attained in the target frequency range by the designed liner. Additionally, the impedance of the uniform HREN-based liner is extracted at flow condition. In all, the…

This paper reviews the nature of substrate vibration within aquatic environments where seismic interface waves may travel along the surface of the substrate, generating high levels of particle motion. There are, however, few data on the ambient levels of particle motion close to the seabed and within the substrates of lakes and rivers. Nor is there information on the levels and the characteristics of the particle motion generated by anthropogenic sources in and on the substrate, which may have major effects upon fishes and invertebrates, all of which primarily detect particle motion. We therefore consider how to monitor substrate vibration and describe the information gained from modeling it. Unlike most acoustic modeling, we treat the substrate as a solid. Furthermore, we use a model where the substrate stiffness increases with depth but makes use of a wave that propagates with little or no dispersion. This shows the presence of higher levels of particle motion than those predicted from the acoustic pressures, and we consider the possible effects of substrate vibration upon fishes…

This study examines the use of Gaussian process (GP) regression for sound field reconstruction. GPs enable the reconstruction of a sound field from a limited set of observations based on the use of a covariance function (a kernel) that models the spatial correlation between points in the sound field. Significantly, the approach makes it possible to quantify the uncertainty on the reconstruction in a closed form. In this study, the relation between reconstruction based on GPs and classical reconstruction methods based on linear regression is examined from an acoustical perspective. Several kernels are analyzed for their potential in sound field reconstruction, and a hierarchical Bayesian parameterization is introduced, which enables the construction of a plane wave kernel of variable sparsity. The performance of the kernels is numerically studied and compared to classical reconstruction methods based on linear regression. The results demonstrate the benefits of using GPs in sound field analysis. The hierarchical parameterization shows the overall best performance, adequately…

In this paper, an experimental characterisation of low Reynolds number rotors is performed in an anechoic room. Two commercially available two-bladed rotors as well as four three-dimensional (3D)-printed rotors with different numbers of blades (from two to five) are tested. The latter have canonical geometry, with an NACA0012 blade section profile, extruded in the radial direction with constant chord and constant 10° pitch. The experimental setup and the 3D printing strategy are first validated using results from the literature for the commercially available rotors. For all the tested rotors, four noise characteristics are analysed: the overall sound pressure level (OASPL), the amplitude of the blade passing frequency (BPF), and the amplitude of its first harmonic and the high-frequency broadband noise. For all the rotors, an increase in all noise characteristics is observed with the rotational speed (rpm) for all directivity angles. Moreover, an interesting change of pattern is observed for the amplitudes of the BPF and of its first harmonic, with, in the vicinity of the rotor…

An empirical model for wind-generated underwater noise is presented that was developed using an extensive dataset of acoustic field recordings and a global wind model. These data encompass more than one hundred years of recording-time and capture high wind events, and were collected both on shallow continental shelves and in open ocean deep-water settings. The model aims to explicitly separate noise generated by wind-related sources from noise produced by anthropogenic sources. Two key wind-related sound-generating mechanisms considered are: surface wave and turbulence interactions, and bubble and bubble cloud oscillations. The model for wind-generated noise shows small frequency dependence (5 dB/decade) at low frequencies (10-100 Hz), and larger frequency dependence (∼15 dB/decade) at higher frequencies (400 Hz-20 kHz). The relationship between noise level and wind speed is linear for low wind speeds (<3.3 m/s) and increases to a higher power law (two or three) at higher wind speeds, suggesting a transition between surface wave/turbulence and bubble source mechanisms. At the…

A deep transfer learning (DTL) method is proposed for the direction of arrival (DOA) estimation using a single-vector sensor. The method involves training of a convolutional neural network (CNN) with synthetic data in source domain and then adapting the source domain to target domain with available at-sea data. The CNN is fed with the cross-spectrum of acoustical pressure and particle velocity during the training process to learn DOAs of a moving surface ship. For domain adaptation, first convolutional layers of the pre-trained CNN are copied to a target CNN, and the remaining layers of the target CNN are randomly initialized and trained on at-sea data. Numerical tests and real data results suggest that the DTL yields more reliable DOA estimates than a conventional CNN, especially with interfering sources.

In the current pandemic, lung ultrasound (LUS) played a useful role in evaluating patients affected by COVID-19. However, LUS remains limited to the visual inspection of ultrasound data, thus negatively affecting the reliability and reproducibility of the findings. Moreover, many different imaging protocols have been proposed, most of which lacked proper clinical validation. To address these problems, we were the first to propose a standardized imaging protocol and scoring system. Next, we developed the first deep learning (DL) algorithms capable of evaluating LUS videos providing, for each video-frame, the score as well as semantic segmentation. Moreover, we have analyzed the impact of different imaging protocols and demonstrated the prognostic value of our approach. In this work, we report on the level of agreement between the DL and LUS experts, when evaluating LUS data. The results show a percentage of agreement between DL and LUS experts of 85.96% in the stratification between patients at high risk of clinical worsening and patients at low risk. These encouraging results…

Although clinical use of the auditory brainstem response (ABR) to detect retrocochlear disorders has been largely replaced by imaging in recent years, the discovery of cochlear synaptopathy has thrown this foundational measure of auditory function back into the spotlight. Whereas modern imaging now allows for the noninvasive detection of vestibular schwannomas, imaging technology is not currently capable of detecting cochlear synaptopathy, the loss of the synaptic connections between the inner hair cells and afferent auditory nerve fibers. However, animal models indicate that the amplitude of the first wave of the ABR, a far-field evoked potential generated by the synchronous firing of auditory nerve fibers, is highly correlated with synaptic integrity. This has led to many studies investigating the use of the ABR as a metric of synaptopathy in humans. However, these studies have yielded mixed results, leading to a lack of consensus about the utility of the ABR as an indicator of synaptopathy. This review summarizes the animal and human studies that have investigated the ABR as a…

This paper is part of a special issue on machine learning in acoustics. A model-based convolutional neural network (CNN) approach is presented to test the viability of this method as an alternative to conventional matched-field processing (MFP) for underwater source-range estimation. The networks are trained with simulated data generated under a particular model of the environment. When tested with data simulated in environments that deviate slightly from the training environment, this approach shows improved prediction accuracy and lower mean-absolute-error (MAE) compared to MFP. The performance of this model-based approach also transfers to real data, as demonstrated separately with field data collected in the Beaufort Sea and off the coast of Southern California. For the former, the CNN predictions are consistent with expected source range while for the latter, the CNN estimates have lower MAE compared to MFP. Examination of the trained CNNs' intermediate outputs suggests that the approach is more constrained than MFP from outputting very inaccurate predictions when there is a…

Though necessary, protective mask wearing in response to the COVID-19 pandemic presents communication challenges. The present study examines how signal degradation and loss of visual information due to masks affects intelligibility and memory for native and non-native speech. We also test whether clear speech can alleviate perceptual difficulty for masked speech. One native and one non-native speaker of English recorded video clips in conversational speech without a mask and conversational and clear speech with a mask. Native English listeners watched video clips presented in quiet or mixed with competing speech. The results showed that word recognition and recall of speech produced with a mask can be as accurate as without a mask in optimal listening conditions. Masks affected non-native speech processing at easier noise levels than native speech. Clear speech with a mask significantly improved accuracy in all listening conditions. Speaking clearly, reducing noise, and using surgical masks as well as good signal amplification can help compensate for the loss of intelligibility due…

Offshore wind farms are part of the transition to a sustainable energy supply and both the total numbers and size of wind turbines are rapidly increasing. While the impact of underwater sound related to construction work has been in the focus of research and regulation, few data exist on the potential impact of underwater sound from operational wind farms. Here, we reviewed published sound levels of underwater sound from operational wind farms and found an increase with size of wind turbines expressed in terms of their nominal power. This trend was identified in both broadband and turbine-specific spectral band sound pressure levels (SPLs). For a nominal power of 10 MW, the trends in broadband SPLs and turbine-specific spectral band SPLs yielded source levels of 170 and 177 dB re 1 μPa m, respectively. The shift from using gear boxes to direct drive technology is expected to reduce the sound level by 10 dB. Using the National Oceanic Atmospheric Administration criterion for behavioral disruption for continuous noise (i.e., level B), a single 10 MW direct drive turbine is expected to…

Face-to-face speech data collection has been next to impossible globally as a result of the COVID-19 restrictions. To address this problem, simultaneous recordings of three repetitions of the cardinal vowels were made using a Zoom H6 Handy Recorder with an external microphone (henceforth, H6) and compared with two alternatives accessible to potential participants at home: the Zoom meeting application (henceforth, Zoom) and two lossless mobile phone applications (Awesome Voice Recorder, and Recorder; henceforth, Phone). F0 was tracked accurately by all of the devices; however, for formant analysis (F1, F2, F3), Phone performed better than Zoom, i.e., more similarly to H6, although the data extraction method (VoiceSauce, Praat) also resulted in differences. In addition, Zoom recordings exhibited unexpected drops in intensity. The results suggest that lossless format phone recordings present a viable option for at least some phonetic studies.

Unmanned aerial vehicle (UAV) technologies are rapidly advancing due to the unlimited number of applications from parcel delivery to people transportation. As the UAV market expands, community noise impact will become a significant problem for public acceptance. Compact drone architectures based on contra-rotating propellers bring significant benefits in terms of aerodynamic performance and redundancy to ensure vehicle control in case of component failure. However, contra-rotating propellers are severely noisy if not designed appropriately. In the framework of a perception-influenced design approach, this paper investigates the optimal rotor spacing distance configuration to minimise noise annoyance. On the basis of a series of psychoacoustic metrics (i.e., loudness, fluctuation strength, roughness, sharpness, and tonality) and psychoacoustic annoyance (PA) models, the optimal rotor axial separation distance (expressed as a function of propeller blade diameter) is at a range from 0.2 to 0.4. This paper also discusses the performance of currently available psychoacoustic models to…

COVID-19 is a global health crisis that has been affecting our daily lives throughout the past year. The symptomatology of COVID-19 is heterogeneous with a severity continuum. Many symptoms are related to pathological changes in the vocal system, leading to the assumption that COVID-19 may also affect voice production. For the first time, the present study investigates voice acoustic correlates of a COVID-19 infection based on a comprehensive acoustic parameter set. We compare 88 acoustic features extracted from recordings of the vowels /i:/, /e:/, /u:/, /o:/, and /a:/ produced by 11 symptomatic COVID-19 positive and 11 COVID-19 negative German-speaking participants. We employ the Mann-Whitney U test and calculate effect sizes to identify features with prominent group differences. The mean voiced segment length and the number of voiced segments per second yield the most important differences across all vowels indicating discontinuities in the pulmonic airstream during phonation in COVID-19 positive participants. Group differences in front vowels are additionally reflected in…

Passive acoustic monitoring has proven to be an indispensable tool for many aspects of baleen whale research. Manual detection of whale calls on these large data sets demands extensive manual labor. Automated whale call detectors offer a more efficient approach and have been developed for many species and call types. However, calls with a large level of variability such as fin whale (Balaenoptera physalus) 40 Hz call and blue whale (B. musculus) D call have been challenging to detect automatically and hence no practical automated detector exists for these two call types. Using a modular approach consisting of faster region-based convolutional neural network followed by a convolutional neural network, we have created automated detectors for 40 Hz calls and D calls. Both detectors were tested on recordings with high- and low density of calls and, when selecting for detections with high classification scores, they were shown to have precision ranging from 54% to 57% with recall ranging from 72% to 78% for 40 Hz and precision ranging from 62% to 64% with recall ranging from 70 to 73%…

This paper presents a semi-analytical method of suppressing acoustic scattering using reinforcement learning (RL) algorithms. We give a RL agent control over design parameters of a planar configuration of cylindrical scatterers in water. These design parameters control the position and radius of the scatterers. As these cylinders encounter an incident acoustic wave, the scattering pattern is described by a function called total scattering cross section (TSCS). Through evaluating the gradients of TSCS and other information about the state of the configuration, the RL agent perturbatively adjusts design parameters, considering multiple scattering between the scatterers. As each adjustment is made, the RL agent receives a reward negatively proportional to the root mean square of the TSCS across a range of wavenumbers. Through maximizing its reward per episode, the agent discovers designs with low scattering. Specifically, the double deep Q-learning network and the deep deterministic policy gradient algorithms are employed in our models. Designs discovered by the RL algorithms performed…

The unprecedented lockdowns resulting from COVID-19 in spring 2020 triggered changes in human activities in public spaces. A predictive modeling approach was developed to characterize the changes in the perception of the sound environment when people could not be surveyed. Building on a database of soundscape questionnaires (N = 1,136) and binaural recordings (N = 687) collected in 13 locations across London and Venice during 2019, new recordings (N = 571) were made in the same locations during the 2020 lockdowns. Using these 30-s-long recordings, linear multilevel models were developed to predict the soundscape pleasantness (R2=0.85) and eventfulness (R2=0.715) during the lockdown and compare the changes for each location. The performance was above average for comparable models. An online listening study also investigated the change in the sound sources within the spaces. Results indicate (1) human sounds were less dominant and natural sounds more dominant across all locations; (2) contextual information is important for predicting pleasantness but not for eventfulness; (3)…

In this study we investigated the effect of sex- and age-related differences in vocal fold length, thickness, and depth on voice production in a three-dimensional vocal fold model. The results showed that the cause-effect relationships between vocal fold physiology and voice production previously identified in an adult male-like vocal fold geometry remained qualitatively the same in vocal folds with geometry representative of adult females and children. We further showed that the often-observed differences in voice production between adult males, adult females, and children can be explained by differences in length and thickness. The lower F0, higher flow rate, larger vocal fold vibration amplitude, and higher sound pressure level (SPL) in adult males as compared to adult females and children can be explained by differences in vocal fold length. In contrast, the thickness effect dominated and contributed to the larger closed quotient of vocal fold vibration, larger normalized maximum flow declination rate, and lower H1-H2 in adult males as compared to adult females and children. The…

The COVID-19 outbreak was announced as a global pandemic by the World Health Organization in March 2020 and has affected a growing number of people in the past few months. In this context, advanced artificial intelligence techniques are brought to the forefront as a response to the ongoing fight toward reducing the impact of this global health crisis. In this study, potential use-cases of intelligent speech analysis for COVID-19 identification are being developed. By analyzing speech recordings from COVID-19 positive and negative patients, we constructed audio- and symptomatic-based models to automatically categorize the health state of patients, whether they are COVID-19 positive or not. For this purpose, many acoustic features were established, and various machine learning algorithms are being utilized. Experiments show that an average accuracy of 80% was obtained estimating COVID-19 positive or negative, derived from multiple cough and vowel /a/ recordings, and an average accuracy of 83% was obtained estimating COVID-19 positive or negative patients by evaluating six symptomatic…

Over the past decade, the range of applications in biomedical ultrasound exploiting 3D printing has rapidly expanded. For wavefront shaping specifically, 3D printing has enabled a diverse range of new, low-cost approaches for controlling acoustic fields. These methods rely on accurate knowledge of the bulk acoustic properties of the materials; however, to date, robust knowledge of these parameters is lacking for many materials that are commonly used. In this work, the acoustic properties of eight 3D-printed photopolymer materials were characterised over a frequency range from 1 to 3.5 MHz. The properties measured were the frequency-dependent phase velocity and attenuation, group velocity, signal velocity, and mass density. The materials were fabricated using two separate techniques [PolyJet and stereolithograph (SLA)], and included Agilus30, FLXA9960, FLXA9995, Formlabs Clear, RGDA8625, RGDA8630, VeroClear, and VeroWhite. The range of measured density values across all eight materials was 1120–1180 kg · m−3, while the sound speed values were between 2020 to 2630 m · s−1, and…

This paper presents methods for the estimation of the time-varying directions of arrival (DOAs) of signals emitted by moving sources. Following the sparse Bayesian learning (SBL) framework, prior information of unknown source amplitudes is modeled as a multi-variate Gaussian distribution with zero-mean and time-varying variance parameters. For sequential estimation of the unknown variance, we present two sequential SBL-based methods that propagate statistical information across time to improve DOA estimation performance. The first method heuristically calculates the parameters of an inverse-gamma hyperprior based on the source signal estimate from the previous time step. In addition, a second sequential SBL method is proposed, which performs a prediction step to calculate the prior distribution of the current variance parameter from the variance parameter estimated at the previous time step. The SBL-based sequential processing provides high-resolution DOA tracking capabilities. Performance improvements are demonstrated by using simulated data as well as real data from the SWellEx-96…

Deep clustering was applied to unlabeled, automatically detected signals in a coral reef soundscape to distinguish fish pulse calls from segments of whale song. Deep embedded clustering (DEC) learned latent features and formed classification clusters using fixed-length power spectrograms of the signals. Handpicked spectral and temporal features were also extracted and clustered with Gaussian mixture models (GMM) and conventional clustering. DEC, GMM, and conventional clustering were tested on simulated datasets of fish pulse calls (fish) and whale song units (whale) with randomized bandwidth, duration, and SNR. Both GMM and DEC achieved high accuracy and identified clusters with fish, whale, and overlapping fish and whale signals. Conventional clustering methods had low accuracy in scenarios with unequal-sized clusters or overlapping signals. Fish and whale signals recorded near Hawaii in February-March 2020 were clustered with DEC, GMM, and conventional clustering. DEC features demonstrated the highest accuracy of 77.5% on a small, manually labeled dataset for classifying signals…

We present an effective thermoviscous theory of acoustofluidics including pressure acoustics, thermoviscous boundary layers, and streaming for fluids embedded in elastic cavities. By including thermal fields, we thus extend the effective viscous theory by Bach and Bruus [J. Acoust. Soc. Am. 144, 766 (2018)]. The acoustic temperature field and the thermoviscous boundary layers are incorporated analytically as effective boundary conditions and time-averaged body forces on the thermoacoustic bulk fields. Because it avoids resolving the thin boundary layers, the effective model allows for numerical simulation of both thermoviscous acoustic and time-averaged fields in three-dimensional models of acoustofluidic systems. We show how the acoustic streaming depends strongly on steady and oscillating thermal fields through the temperature dependency of the material parameters, in particular the viscosity and the compressibility, affecting both the boundary conditions and spawning additional body forces in the bulk. We also show how even small steady temperature gradients (∼1 K/mm) induce…

Conceiving lightweight structures with low vibration and sound radiation properties is an important topic. The concept of Acoustic Black Hole (ABH) offers new impetus to tackle this problem. Most existing ABH structures are based on simple ABH cells. Apart from the reduced structural strength, systematic ABH effects occur typically above the cut-on frequency of the ABH element, which is perceived as a bottlenecking problem. To tackle the problem, this paper examines the sound radiation properties of a plate comprising periodically tangled ABH cells. Through combining ABH effects with sub-wavelength bandgaps (BGs), numerical and experimental studies show that the plate exhibits reduced sound radiation properties in an ultra-broad frequency range, far below the cut-on frequency of an ABH element. This is owing to the tangled nature of the ABH elements, which extends the actual dimension of the ABH, lowers its onset frequency and reduces the sound radiation efficiency through creating slow waves. Inside the BGs, the reduced sound radiation is mainly due to the redistribution of the…

Metamaterials are designed by arranging artificial structural elements according to periodic geometries to obtain advantageous and unusual properties when they are hit by waves. Initially designed to interact with electromagnetic waves, their use naturally extended to sound waves, proving to be particularly useful for the construction of containment and soundproofing systems in buildings. In this work, a new metamaterial has been developed with the use of a polyvinyl chloride membrane on which buttons have been glued. Two types of buttons were used, with different weights, placing them on the membrane according to a radial geometry. Each sample of metamaterial was subjected to sound absorption coefficient measurements using the impedance tube. Measurements were made using the samples by setting three configurations, creating a cavity with different thicknesses. The results of the measurements were subsequently used as input for training a simulation model based on artificial neural networks. The model showed an excellent generalization capacity, returning estimates of the acoustic…

The effect of face covering masks on listeners' recall of spoken sentences was investigated. Thirty-two German native listeners watched video recordings of a native speaker producing German sentences with and without a face mask, and then completed a cued-recall task. Listeners recalled significantly fewer words when the sentences had been spoken with a face mask. This might suggest that face masks increase processing demands, which in turn leaves fewer resources for encoding speech in memory. The result is also informative for policy-makers during the COVID-19 pandemic, regarding the impact of face masks on oral communication.

Differences in speakers' voice characteristics, such as mean fundamental frequency (F0) and vocal-tract length (VTL), that primarily define speakers' so-called perceived voice gender facilitate the perception of speech in competing speech. Perceiving speech in competing speech is particularly challenging for children, which may relate to their lower sensitivity to differences in voice characteristics than adults. This study investigated the development of the benefit from F0 and VTL differences in school-age children (4-12 years) for separating two competing speakers while tasked with comprehending one of them and also the relationship between this benefit and their corresponding voice discrimination thresholds. Children benefited from differences in F0, VTL, or both cues at all ages tested. This benefit proportionally remained the same across age, although overall accuracy continued to differ from that of adults. Additionally, children's benefit from F0 and VTL differences and their overall accuracy were not related to their discrimination thresholds. Hence, although children's…

It has been argued that the relative position of spectral envelopes along the frequency axis serves as a cue for musical instrument size (e.g., violin vs viola) and that the shape of the spectral envelope encodes family identity (violin vs flute). It is further known that fundamental frequency (F0), F0-register for specific instruments, and dynamic level strongly affect spectral properties of acoustical instrument sounds. However, the associations between these factors have not been rigorously quantified for a representative set of musical instruments. Here, we analyzed 5640 sounds from 50 sustained orchestral instruments sampled across their entire range of F0s at three dynamic levels. Regression of spectral centroid (SC) values that index envelope position indicated that smaller instruments possessed higher SC values for a majority of instrument classes (families), but SC also correlated with F0 and was strongly and consistently affected by the dynamic level. Instrument classification using relatively low-dimensional cepstral audio descriptors allowed for discrimination between…

The widespread use of face coverings during the COVID-19 pandemic has created communication challenges for many individuals, particularly for those who are deaf or hard of hearing and for those who must speak through masks in suboptimal conditions. This study includes some newer mask options as well as transparent masks to help those who depend on lipreading and other facial cues. The results corroborate earlier published results for non-transparent masks, but transparent options have greater attenuation, resonant peaks, and deflect sounds in ways that non-transparent masks do not. Although transparent face coverings have poorer acoustic performance, the presence of visual cues remains important for both verbal and non-verbal communication. Fortunately, there are creative solutions and technologies available to overcome audio and/or visual barriers caused by face coverings.

This paper proposes a robust system for detecting North Atlantic right whales by using deep learning methods to denoise noisy recordings. Passive acoustic recordings of right whale vocalisations are subject to noise contamination from many sources, such as shipping and offshore activities. When such data are applied to uncompensated classifiers, accuracy falls substantially. To build robustness into the detection process, two separate approaches that have proved successful for image denoising are considered. Specifically, a denoising convolutional neural network and a denoising autoencoder, each of which is applied to spectrogram representations of the noisy audio signal, are developed. Performance is improved further by matching the classifier training to include the vestigial signal that remains in clean estimates after the denoising process. Evaluations are performed first by adding white, tanker, trawler, and shot noises at signal-to-noise ratios from -10 to +5 dB to clean recordings to simulate noisy conditions. Experiments show that denoising gives substantial improvements to…

Two main methods have been proposed to derive the acoustical radiation force and torque applied by an arbitrary acoustic field on a particle: The first one relies on the plane wave angular spectrum decomposition of the incident field (see Sapozhnikov and Bailey [J. Acoust. Soc. Am. 133, 661-676 (2013)] for the force and Gong and Baudoin [J. Acoust. Soc. Am. 148, 3131-3140 (2020)] for the torque), while the second one relies on the decomposition of the incident field into a sum of spherical waves, the so-called multipole expansion (see Silva [J. Acoust. Soc. Am. 130, 3541-3544 (2011)] and Baresch, Thomas, and Marchiano [J. Acoust. Soc. Am. 133, 25-36 (2013)] for the force, and Silva, Lobo, and Mitri [Europhys. Lett. 97, 54003 (2012)] and Gong, Marston, and Li [Phys. Rev. Appl. 11, 064022 (2019)] for the torque). In this paper, we formally establish the equivalence between the expressions obtained with these two methods for both the force and torque.

Acoustic holographic techniques are crucial in diverse applications, such as three-dimensional holographic display and particle manipulation. However, conventional methods for computer-generated acoustics holography rely heavily on iterative optimization algorithms, which are time-consuming and particularly hinder their capacity of generating a dynamic hologram in real time. Here, a deep learning approach based on U-Net is proposed to rapidly generate an acoustic hologram with optimal amplitude and phase maps. It is demonstrated that, after being trained with adequate data that are numerically synthesized by the pseudo-inverse method, the proposed deep learning approach can generate both amplitude and phase maps for new target images with an improved overall reconstruction quality. Remarkably, after the offline cost is compensated by a lower online cost for the proposed DL approach, the hologram generation speed is significantly accelerated by the proposed deep learning approach as compared with the pseudo-inverse method, especially for complicated or dynamic images. With the…

The near and far fields of traditional loudspeakers are differentiated by whether the sound pressure amplitude is inversely proportional to the propagating distance. However, the audio sound field generated by a parametric array loudspeaker (PAL) is more complicated, and in this article it is proposed to be divided into three regions: near field, Westervelt far field, and inverse-law far field. In the near field, the audio sound experiences strong local effects and an efficient quasilinear solution is presented. In the Westervelt far field, local effects are negligible so that the Westervelt equation is used, and in the inverse-law far field, a simpler solution is adopted. It is found that the boundary between the near and Westervelt far fields for audio sound lies at approximately a2/λ – λ/4, where a is transducer radius and λ is ultrasonic wavelength. At large transducer radii and high ultrasonic frequencies, the boundary moves close to the PAL and can be estimated by a closed-form formula. The inverse-law holds for audio sound in the inverse-law far field and is more than 10…

When the COVID-19 pandemic halted in-person data collection, many linguists adopted new online technologies to replace traditional methods, including video conferencing applications (apps) like Zoom (Zoom Video Communications, San Jose, CA), which allow live interaction with remote participants. This study evaluated the suitability of video calls for the phonetic analysis of vowel configurations, mergers, and nasalization by comparing simultaneous recordings from three popular video conferencing apps (Zoom; Microsoft Skype, Redmond, WA; Microsoft Teams, Redmond, WA) to those taken from professional equipment (H4n field recorder) and an offline iPad (Apple, Cupertino, CA) identical to those running the apps. All three apps conveyed vowel arrangements and nasalization patterns relatively faithfully, but absolute measurements varied, particularly for the female speaker and in the 750-1500 Hz range, which affected the locations (F1 × F2) of low and back vowels and reduced nasalization measurements (A1-P0) for the female's prenasal vowels. Based on these results, we assess the validity…

While source localization and seabed classification are often approached separately, the convolutional neural networks (CNNs) in this paper simultaneously predict seabed type, source depth and speed, and the closest point of approach. Different CNN architectures are applied to mid-frequency tonal levels from a moving source recorded on a 16-channel vertical line array (VLA). After training each CNN on synthetic data, a statistical representation of predictions on test cases is presented. The performance of a single regression-based CNN is compared to a multitask CNN in which regression is used for the source parameters and classification for the seabed type. The impact of water sound speed profile and seabed variations on the predictions is evaluated using simulated test cases. Environmental mismatch between the training and testing data has a negative impact on source depth estimates, while the remaining labels are estimated tolerably well but with a bias towards shorter ranges. Similar results are found for data measured on two VLAs during Seabed Characterization Experiment 2017.…

No agreed-upon method currently exists for objective measurement of perceived voice quality. This paper describes validation of a psychoacoustic model designed to fill this gap. This model includes parameters to characterize the harmonic and inharmonic voice sources, vocal tract transfer function, fundamental frequency, and amplitude of the voice, which together serve to completely quantify the integral sound of a target voice sample. In experiment 1, 200 voices with and without diagnosed vocal pathology were fit with the model using analysis-by-synthesis. The resulting synthetic voice samples were not distinguishable from the original voice tokens, suggesting that the model has all the parameters it needs to fully quantify voice quality. In experiment 2 parameters that model the harmonic voice source were removed one by one, and the voice tokens were re-synthesized with the reduced model. In every case the lower-dimensional models provided worse perceptual matches to the quality of the natural tokens than did the original set, indicating that the psychoacoustic model cannot be…

We present a theoretical framework for acoustic wave propagation in a metasurface comprising a hexagonal lattice of hard spherical inclusions embedded in a soft elastic medium. Each layer of inclusions in the direction of sound propagation is approximated as a homogenized layer with effective geometric and material properties. To account for multiple scattering effects in the lattice of resonant inclusions, an analogy between the fluid dynamics of creeping flows and elastodynamics of soft materials is implemented. Results obtained analytically are in excellent agreement with numerical simulations that exactly model the geometric and material properties of the metasurface.

Detecting small defects in curved parts through classical monostatic pulse-echo ultrasonic imaging is known to be a challenge. Hence, a robot-assisted ultrasonic testing system with the track-scan imaging method is studied to improve the detecting coverage and contrast of ultrasonic images. To further improve the image resolution, we propose a visual geometry group-UNet (VGG-UNet) deep learning network to optimize the ultrasonic images reconstructed by the track-scan imaging method. The VGG-UNet uses VGG to extract advanced information from ultrasonic images and takes advantage of UNet for small dataset segmentation. A comparison of the reconstructed images on the simulation dataset with ground truth reveals that the peak signal-to-noise ratio (PSNR) and structural similarity index measure (SSIM) can reach 39 dB and 0.99, respectively. Meanwhile, the trained network is also robust against the noise and environmental factors according to experimental results. The experiments indicate that the PSNR and SSIM can reach 32 dB and 0.99, respectively. The resolution of ultrasonic images…

One of the main fields of application of personal sound zones (PSZs) is the car industry. PSZ systems are usually designed assuming that the acoustic environment is not changing. However, due to variability in the car cabin's acoustic conditions, it is necessary to design a system capable of adapting to these variations. This paper introduces the use of the filtered-x least-mean-squares algorithm to adapt a PSZ system in a car cabin when the seat positions change. The article focuses on four aspects. First, the effect of the initial coefficients is investigated to help with the convergence of the adaptive algorithm. Second, a study is made on the definition of the desired pressure and its effect on the reproduction. Third, the system's sensitivity to external noise is also evaluated. Last, the possibility to reduce the number of microphones for the adaptive system is considered. The method's performance is evaluated using a two headrest system installed in a car cabin and when moving the seats to different positions. It is shown that the system is capable of maintaining the…

We introduce a model that describes spherical oscillations of encapsulated microbubbles in an unbounded surrounding fluid. A Rayleigh-Plesset-like equation is derived by coupling the Navier-Stokes equation that describes fluid dynamics with the Navier equation that describes solid dynamics via the internal/external boundary conditions. While previous models were restricted to incompressible isotropic shells, the solid shell is modeled here as a compressible viscoelastic isotropic material and then generalized to an anisotropic material. The exact value of the resonance frequency is calculated analytically, and the damping constant is computed in the approximation of weak damping. A correction of the widely used Church model for incompressible shells is evidenced, and the effects of shell compressibility and anisotropy are discussed.

In this paper, a method for the problem of depth estimation of a broadband source via reliable acoustic path propagation is presented for the case using a vertical line array (VLA). The estimates are determined by two kinds of multipath time-delay ratios, namely, the ratio of direct-surface-reflected (D-SR) to direct-direct time-delays and the ratio of D-SR to surface-reflected-surface-reflected time-delays. The innovation of ratio behavior is that it provides a mechanism for obtaining a useful depth interval with the assumption of plane-wave propagation. The estimation accuracy of a depth interval relies on the degree to which the actual acoustic propagation characteristic can be modeled by image theory. Furthermore, the variability of depth interval due to the approximation made in the derivation method allows one to achieve binary discrimination of both the source depth and source range with only a minimal amount of prior environmental knowledge. The methodology of multipath time-delay estimation is first reviewed and improved, followed by an illustration of the source depth…

Resolvent analysis has demonstrated encouraging results for modeling coherent structures in jets when compared against their data-educed counterparts from high-fidelity large-eddy simulations (LES). We formulate resolvent analysis as an acoustic analogy that relates the near-field resolvent forcing to the near- and far-field pressure. We use an LES database of round, isothermal, Mach 0.9 and 1.5 jets to produce an ensemble of realizations for the acoustic field that we project onto a limited set of resolvent modes. In the near-field, we perform projections on a restricted acoustic output domain, r/D=[5,6], while the far-field projections are performed on a Kirchhoff surface comprising a 100-diameter arc centered at the nozzle. This allows the LES realizations to be expressed in the resolvent basis via a data-deduced, low-rank, cross-spectral density matrix. We find that a single resolvent mode reconstructs the most energetic regions of the acoustic field across Strouhal numbers, St=[0-1], and azimuthal wavenumbers, m=[0,2]. Finally, we present a simple function that results in a…

A finite-element model is presented for numerical simulation in three dimensions of acoustophoresis of suspended microparticles in a microchannel embedded in a polymer chip and driven by an attached piezoelectric transducer at MHz frequencies. In accordance with the recently introduced principle of whole-system ultrasound resonances, an optimal resonance mode is identified that is related to an acoustic resonance of the combined transducer-chip-channel system and not to the conventional pressure half-wave resonance of the microchannel. The acoustophoretic action in the microchannel is of comparable quality and strength to conventional silicon-glass or pure glass devices. The numerical predictions are validated by acoustic focusing experiments on 5-μm-diameter polystyrene particles suspended inside a microchannel, which was milled into a polymethylmethacrylate chip. The system was driven anti-symmetrically by a piezoelectric transducer, driven by a 30-V peak-to-peak alternating voltage in the range from 0.5 to 2.5 MHz, leading to acoustic energy densities of 13 J/m3 and particle…

This work develops a second-order approximation (SOA) model and a three-dimensional (3D) finite element (FE) model to calculate scattering-induced attenuation for elastic wave propagation in polycrystals with elongated grains of arbitrary crystal symmetry. The SOA model accounts for some degree of multiple scattering, whereas the 3D FE model includes all scattering possibilities. The SOA model incorporates the accurate geometric two-point correlation function obtained from the FE material systems to enable comparative studies between the two models. Also, the analytical Rayleigh and stochastic asymptotes are presented to provide explicit insights into propagation behaviors. Quantitative agreement is found between the FE and analytical models for all evaluated cases. In particular, the FE simulations support the SOA model prediction that grain shape does not exert influence on attenuation in the Rayleigh regime and its effect emerges as frequency increases to the stochastic regime showing anisotropy in attenuation. This attenuation anisotropy intensifies with the increase in…

This study employs nonlinear ultrasonic techniques to track microstructural changes in additively manufactured metals. The second harmonic generation technique based on the transmission of Rayleigh surface waves is used to measure the acoustic nonlinearity parameter, β. Stainless steel specimens are made through three procedures: traditional wrought manufacturing, laser-powder bed fusion, and laser engineered net shaping. The β parameter is measured through successive steps of an annealing heat treatment intended to decrease dislocation density. Dislocation density is known to be sensitive to manufacturing variables. In agreement with fundamental material models for the dislocation-acoustic nonlinearity relationship in the second harmonic generation, β drops in each specimen throughout the heat treatment before recrystallization. Geometrically necessary dislocations (GNDs) are measured from electron back-scatter diffraction as a quantitative indicator of dislocations; average GND density and β are found to have a statistical correlation coefficient of 0.852 showing the sensitivity…

Sound field analysis methods make it possible to characterize and reconstruct a sound field from a limited set of observations. Classical approaches rely on the use of analytical basis functions to model the sound field throughout the observed domain. When the complexity of the sound field is high, for example, in a room at mid and high frequencies, propagating wave representations can be suboptimal due to model discrepancy. We examine the use of local representations to alleviate this model discrepancy and explore data-driven approaches to obtain suitable models. Specifically, local representations are used to reconstruct the sound field over a large spatial aperture in a room. The performance of local models is compared against conventional plane wave reconstructions and the use of data-driven local functions is examined. Dictionary learning and principal component analysis are used to obtain functions from extensive spatial measurements in an empty room. The results indicate that local partitioning models conform to fields of high spatial complexity. Dictionary learning…

This project acquired sound levels logged across six school days and impulse responses in 220 classrooms across four K-12 grades. Seventy-four percent met reverberation time recommendations. Sound levels were processed to estimate occupied signal-to-noise ratios (SNRs), using Gaussian mixture modeling and from daily equivalent and statistical levels. A third method, k-means clustering, estimated SNR more precisely, separating data on nine dimensions into one group with high levels across speech frequencies and one without. The SNRs calculated as the daily difference between the average levels for the speech and non-speech clusters are found to be lower than 15 dB in 27.3% of the classrooms and differ from using the other two methods. The k-means data additionally indicate that speech occurred 30.5%-81.2% of the day, with statistically larger percentages found in grade 3 compared to higher grades. Speech levels exceeded 65 dBA 35% of the day, and non-speech levels exceeded 50 dBA 32% of the day, on average, with grades 3 and 8 experiencing speech levels exceeding 65 dBA statistically…

The goal of this project is to use acoustic signatures to detect, classify, and count the calls of four acoustic populations of blue whales so that, ultimately, the conservation status of each population can be better assessed. We used manual annotations from 350 h of audio recordings from the underwater hydrophones in the Indian Ocean to build a deep learning model to detect, classify, and count the calls from four acoustic song types. The method we used was Siamese neural networks (SNN), a class of neural network architectures that are used to find the similarity of the inputs by comparing their feature vectors, finding that they outperformed the more widely used convolutional neural networks (CNN). Specifically, the SNN outperform a CNN with 2% accuracy improvement in population classification and 1.7%-6.4% accuracy improvement in call count estimation for each blue whale population. In addition, even though we treat the call count estimation problem as a classification task and encode the number of calls in each spectrogram as a categorical variable, SNN surprisingly learned the…