Sound Recordings Research Articles

Research in the Department of Music Technology and Acoustics of the Hellenic Mediterranean University: An Overview and Prospects

The Department of Music Technology and Acoustics of the Hellenic Mediterranean University offers a unique higher education program in Greece, addressing the growing demand for specialists in music technology, sound technology, and acoustics. It aims to educate specialized professionals in the rapidly advancing scientific fields of music technology and acoustics, mainly driven by the swift progress in electronic technology. The Department aims to address a gap in the professional market by producing highly skilled graduates, capable not only of keeping up with the latest scientific and technological developments but also of leading the way by introducing innovative approaches and methods. The Department combines art, science, and technology, focusing on sound recording, analysis, synthesis, and music production. Music technology encompasses various cutting-edge fields such as network music performance, artificial intelligence in music, and music embodiment. Acoustics refers to fundamental aspects of sound as well as its generation, transmission, and related phenomena. It includes research fields such as physical acoustics, optoacoustics, and vibroacoustics. This overview presents the research activities, methodologies, and results. A discussion of future research works and pointers to future technological evolution towards real-world music and acoustics applications is also provided.

Impact of the Passage of Time on the Correct Identification of the Speaker Using the Auditory Method

Courts in Poland, as well as in most countries in the world, allow for the identification of a person on the basis of his/her voice using the so-called voice presentation method, i.e., the auditory method. This method is used in situations where there is no sound recording and the perpetrator of the criminal act was masked and the victim heard only his or her voice. However, psychologists, forensic acousticians, as well as researchers in the field of auditory perception and forensic science more broadly describe many cases in which such testimony resulted in misjudgement. This paper presents the results of an experiment designed to investigate, in a Polish language setting, the extent to which the passage of time impairs the correct identification of a person. The study showed that 31 days after the speaker’s voice was first heard, the correct identification for a female voice was 30% and for a male voice 40%.

Analysis of Vocal Signatures of COVID-19 in Cough Sounds: A Newer Diagnostic Approach Using Artificial Intelligence.

Artificial intelligence (AI) based models are explored increasingly in the medical field. The highly contagious pandemic of coronavirus disease 2019 (COVID-19) affected the world and availability of diagnostic tools high resolution computed tomography (HRCT) and/or real-time reverse transcriptase-polymerase chain reaction (RTPCR) was very limited, costly and time consuming. Therefore, the use of AI in COVID-19 for diagnosis using cough sounds can be efficacious and cost effective for screening in clinic or hospital and help in early diagnosis and further management of patients. To develop an accurate and fast voice-processing AI software to determine voice-based signatures in discriminating COVID-19 and non-COVID-19 cough sounds for screening of COVID-19. A prospective study involving 117 patients was performed based on online and/or offline voice data collection of cough sounds of COVID-19 patients in isolation ward of a tertiary care teaching hospital and non-COVID-19 participants using a smart phone. A website-based AI software was developed to identify the cough sounds as COVID-19 or non-COVID-19. The data were divided into three segments including training set, validation set and test set. A pre-processing algorithmwas utilized and combined with Short Time Fourier Transform feature representation and Logistic regression model. A precise software was used to identify vocal signatures and K-fold cross validation was carried out. A total of 117 audio recordings of cough sounds were collected through the developed website after inclusion-exclusion criteria out of which 52 have been marked belonging to COVID-19 positive, while 65 were marked as COVID-19 negative/unsure /never had COVID-19, which were assumed to be COVID-19 negative based on RT-PCR test results. The mean and standard error values for the accuracies attained at the end of each experiment in training, validation and testing set were found to be 67.34%±0.22, 58.57%±1.11 and 64.60%±1.79 respectively. The weight values were found to be positive which were contributing towards predicting the samples as COVID-19 positive with large spikes around 7.5 kHz, 7.8 kHz, 8.6 kHz and 11 kHz which can be used for classification. The proposed AI based approach can be a helpful screening tool for COVID-19 using vocal sounds of cough. It can help the health system by reducing the cost burden and improving overall diagnosis and management of the disease.

A price on their heads? Assessing foreign demand as a driver of hornbill hunting in Cameroon

Unsustainable hunting to supply wildlife markets is a global threat to bird species and a key driver of biodiversity declines across Africa. Despite well-documented negative impacts of wildlife trade on many bird species, such as Asian hornbills and African vultures, its consequences for African hornbills remain poorly understood. Here, we use data from both online trade and interviews with hunters in Cameroon to investigate the implications of foreign demand for African hornbills, using a twofold approach. First, we explored factors associated with the prices of hornbill skulls, the most heavily traded and expensive hornbill products, in online markets. Second, we examined the relationship between hunting pressure and market demand using semi-structured interviews with hornbill hunters and traders in Cameroon. In our online sales analysis, we considered factors including numbers of sales, species conservation status, Convention on International Trade in Endangered Species of Wild Fauna and Flora (CITES) listing, body size, sex, and skull length (including bill and casque). In addition, we considered the number of sound recordings available for focal species, as this factor is highly correlated with species range size and may serve as a proxy for detectability in the field. We then used generalized linear mixed-effects models (GLMM) to identify variables closely associated with the prices of hornbill skulls online. Generalized linear models (GLM) were used to examine the relationship between hunting pressure and the local and foreign demands in Cameroon. We found that online trade included skulls and other products from 20 African and five Asian hornbill species, the latter of which were all CITES-listed. The median online price for a hornbill skull was £117. Nearly half (nine out of 20) of the African hornbill species found in online trade were native to Cameroon. The three most heavily traded species—Yellow-casqued Hornbill (Ceratogymna elata), Black-casqued Hornbill (C. atrata), and White-thighed Hornbill (Bycanistes albotibialis)—accounted for 66% of online hornbill products. Larger-bodied hornbill species classified as threatened, subject to CITES regulations, and species with higher numbers of sound recordings available tended to command higher prices online and thus likely face higher hunting pressure. We also found that foreign demand has a significant impact on hornbill hunting pressure, particularly among younger hunters. Approximately 73% of Cameroonian hunters interviewed targeted hornbills, with 91% of these motivated to acquire hornbill heads for trade, and 98% intending to sell their products to foreign buyers. Our findings highlight the urgent need to address the negative impacts of international wildlife trade and implement measures to protect African hornbill populations from increasing threats of overexploitation.

SonicGuard Sensor-A Multichannel Acoustic Sensor for Long-Term Monitoring of Abdominal Sounds Examined through a Qualification Study.

Auscultation is a fundamental diagnostic technique that provides valuable diagnostic information about different parts of the body. With the increasing prevalence of digital stethoscopes and telehealth applications, there is a growing trend towards digitizing the capture of bodily sounds, thereby enabling subsequent analysis using machine learning algorithms. This study introduces the SonicGuard sensor, which is a multichannel acoustic sensor designed for long-term recordings of bodily sounds. We conducted a series of qualification tests, with a specific focus on bowel sounds ranging from controlled experimental environments to phantom measurements and real patient recordings. These tests demonstrate the effectiveness of the proposed sensor setup. The results show that the SonicGuard sensor is comparable to commercially available digital stethoscopes, which are considered the gold standard in the field. This development opens up possibilities for collecting and analyzing bodily sound datasets using machine learning techniques in the future.

Heterogeneous features and deep learning networks fusion-based pest detection, prevention and controlling system using IoT and pest sound analytics in a vast agriculture system

Every country's foundation is its agricultural industry, which drives about half of global economic growth. Accurate farming is crucial for crop evaluation and pest control. Old-fashioned pest detection is unstable and gives inaccurate forecasts. These surveillance tactics are intrusive, expensive, and vulnerable to assumptions. Pest sounds can be caught with minimum effort by utilizing IoT networks. Deep learning algorithms, species distribution range assessments, and nature monitoring can automatically identify and categorize pest sounds. The IoT-driven computerized components in this research's unique pest identification system used machine learning on insect sound recordings. The Butterworth filter, Blackman and Flattop window, Ultraspherical Filter, Rife-Vincent Window, Cosine-Tapered Window, FFT, DFT, STFT, PNCC, RASTA-PLPCC, LSFCC, sound detectors, and PID sensors were employed. The intended study used the HFDLNet for training, testing, and validation, examining 7,200 pest sounds from 72 species to determine their unique characteristics and statistical attributes. The proposed model has 99.87 % accuracy, 99.96 % sensitivity, 99.88 % specificity, 99.96 % recall, 99.93 % F1 score, and 99.98 % precision. This study outperforms Inception-ResNet-v2, FRCNN ResNet-50, Fatser-PestNet, MD-YOLO, YOLOv5m, MAM-IncNet, and Xception. The suggested solution uses IoT networks and pest sound analysis to establish a pest preventive and management strategy and a solar-powered generator to power IoT devices across vast agricultural fields.

Evaluating the potential of bioacoustics in avian migration research by citizen science and weather radar observations.

The study of nocturnal bird migration brings observational challenges because of reduced visibility and observability of birds at night. Remote sensing tools, especially radars, have long been the preferred choice of scientists to study nocturnal migrations. A major downside of these remote sensing tools is the lack of species-level information. With technological advances in recent decades and with improved accessibility and affordability of acoustic tools, sound recordings have steeply increased in popularity. In Europe, there is no exhaustive qualitative and quantitative evaluation of the content of such acoustic databases and therefore the value for migration science and migration-related applications, such as bird collision hazard assessments, is mostly unknown. In the present work we compared migration schedules estimated from citizen science data with quantitative temporal occurrence of species in four years of acoustic recordings. Furthermore, we contrasted acoustic recordings with citizen science observations and weather radar data from one spring and one autumn season to assess the qualitative and quantitative yield of acoustic recordings for migration-related research and applications. Migration intensity estimated from weather radar data correlated best at low levels with acoustic records including all species in spring while in autumn passerine species showed stronger correlation than the entire species composition. Our findings identify a minor number of species whose call records may be eligible for applications derived from acoustics. Especially the highly vocal species Song thrush and Redwing showed relatively good correlations with radar and citizen science migration schedules. Most long-distance passerine migrants and many other migrants were not captured by acoustics and an estimated seasonal average of about 50% of nocturnally migrating passerine populations remained undetected. Overall, the ability of acoustic records to act as a proxy of overall migration dynamics is highly dependent on the migration period and species involved.

Phonological Variation of /r/ in Selected Arabic Dialects

The present study offers a descriptive examination of the Arabic phoneme /r/, renowned for its intricate and diverse nature within the Arabic phonological system. This research investigates the varied manifestations of the phoneme /r/ in different regional dialects, with a specific emphasis on its pronunciation in the initial, middle, and final positions of words. The paper's significance is rooted in its contribution to the comprehension of the Arabic /r/ phoneme, a multifaceted and diverse sound within Arabic phonological systems. The paper illuminates the phonetic variability of the /r/ by analyzing its distinct manifestations across diverse Arabic dialects and regions. This can potentially facilitate language acquisition and instruction. The investigation conducted by the authors pertains to the marked and unmarked characteristics of the phonological /r/ in Arabic. The population under investigation comprises a randomly selected group of Arabic-speaking individuals aged 18 years and older, encompassing both genders and hailing from diverse countries and regions. The study employs a descriptive-analytical methodology, utilizing a unique features matrix and the Audacity Sound Recording and Analysis Application to examine the gathered data. The findings of the research enhance the comprehension of the phonetic variability that typifies the Arabic /r/ phoneme and elucidate its manifestation in diverse contexts and geographical locations. Furthermore, the outcomes of this investigation hold significance for the domain of language education and acquisition, and the techniques utilized were designed to guarantee the precision and dependability of the gathered information.

COVID-19 Detection From Respiratory Sounds With Hierarchical Spectrogram Transformers.

Monitoring of prevalent airborne diseases such as COVID-19 characteristically involves respiratory assessments. While auscultation is a mainstream method for preliminary screening of disease symptoms, its utility is hampered by the need for dedicated hospital visits. Remote monitoring based on recordings of respiratory sounds on portable devices is a promising alternative, which can assist in early assessment of COVID-19 that primarily affects the lower respiratory tract. In this study, we introduce a novel deep learning approach to distinguish patients with COVID-19 from healthy controls given audio recordings of cough or breathing sounds. The proposed approach leverages a novel hierarchical spectrogram transformer (HST) on spectrogram representations of respiratory sounds. HST embodies self-attention mechanisms over local windows in spectrograms, and window size is progressively grown over model stages to capture local to global context. HST is compared against state-of-the-art conventional and deep-learning baselines. Demonstrations on crowd-sourced multi-national datasets indicate that HST outperforms competing methods, achieving over 90% area under the receiver operating characteristic curve (AUC) in detecting COVID-19 cases.

Video clips in physics education

Film recordings – video clips – have been used in the teaching and learning processes for decades. At the very beginning, the dominant role was played by television and professionally prepared recordings on portable media (e.g. LaserDiscs). With the advancement of technology and the emergence of new media, as well as the revolution in access to image and sound recording and processing devices, video clips – including amateur ones – have become the dominant path of contemporary visual communication. The paper is of an overview character and discusses examples of educational use of video in physics education in the past and at present, including TV lectures, popular science programs, dedicated video materials and educational environments, software for video measurement and evaluation of learning outcomes. Various illustrations along with literature references are given to outline potential educational pathways in making use of video clips as didactically valuable medium in teaching - learning physics.

Caller identification and characterization of individual humpback whale acoustic behaviour.

Acoustic recording tags provide fine-scale data linking acoustic signalling with individual behaviour; however, when an animal is in a group, it is challenging to tease apart calls of conspecifics and identify which individuals produce each call. This, in turn, prohibits a robust assessment of individual acoustic behaviour including call rates and silent periods, call bout production within and between individuals, and caller location. To overcome this challenge, we simultaneously instrumented small groups of humpback whales on a western North Atlantic feeding ground with sound and movement recording tags. This approach enabled a comparison of the relative amplitude of each call across individuals to infer caller identity for 97% of calls. We recorded variable call rates across individuals (mean = 23 calls/h) and groups (mean = 55 calls/h). Calls were produced throughout dives, and most calls were produced in bouts with short inter-call intervals of 2.2 s. Most calls received a likely response from a conspecific within 100 s. This caller identification (ID) method facilitates studying both individual- and group-level acoustic behaviour, yielding novel results about the nature of sequence production and vocal exchanges in humpback whale social calls. Future studies can expand on these caller ID methods for understanding intra-group communication across taxa.

Posthumous Sound and the General Imagination

Abstract The 1997 discovery of a fifty-thousand-year-old flute made from the femur of a cave bear, with its intimation of reanimating nonhumans, and the 1977 launch of the Voyager spacecraft carrying an eclectic set of sound recordings intended to be heard in the distant future by nonhuman others: two sonic events that frame the possible meanings of posthumous. Together these examples and others question whether everything audible is already over—the bear's lost life, electronic recording procedures—or indefinitely deferred until an act of listening that may never occur. An ecological address to the problems of making sonic culture at a historical turning point at or beyond terminal risk prompts a politics of the commons grounded in a general imagination (modeled on Marx's general intellect). Against earlier modernist claims for both rationality and its failure, sound cultures enact a drama of melancholy and hope in the ecological continuity of body and world at the moment of their end.

Urban background sound recordings for virtual acoustics under various weather conditions

One of the major challenges in the auralization of complex urban simulations is the presence of ambience, for example background sounds. Omitting them can result in a sterile and implausible impression. In this work we present a dataset of spatially recorded ambient sounds with metadata on meteorological and acoustical parameters. The recordings are ready to be admixed with simulated soundscapes corresponding to various weather conditions. In particular, 28 soundscapes have been recorded at the IHTApark (green space next to the Institute for Hearing Technology and Acoustics) in Aachen (Germany) in winter and spring. They have been segmented into 30-s auralizable snippets. Moreover, the database contains a statistical and psychoacoustical evaluation of the recordings.

Onboard sound emission measurements of unmanned aerial vehicles for psychoacoustic experiments

With the prospect of small UAVs being more commonly used for tasks like surveillance or parcel transport within inhabited areas, demand for noise regulations arises. The noise assessment of UAV operations can be investigated through laboratory listening experiments using auralized sound stimuli. Sound recordings of source signals for dynamic flight operations have to be performed outdoors where measurements with stationary microphones impose uncertainties in backpropagation and suffer from bad signal-to-noise ratio due to the weak source and the presence of ambient sounds. To overcome these difficulties, this contribution presents an approach for acquiring UAV source signals using onboard microphones. A lightweight setup for time synchronous recordings of sound using MEMS microphones, the UAV position and the rotational speed of each rotor was developed and applied to two quadcopters of 0.9 and 6.3 kg mass. Source directivity and propagation effects were added to the source signals to obtain clean sound pressure signals at virtual listener positions. The resulting stimuli were successfully used in a listening experiment on short-term noise annoyance.

Link to the Library of Congress Etching the Human Voice: Exploring the History of Sound Recording Technology

Link to the Library of Congress Etching the Human Voice: Exploring the History of Sound Recording Technology

TRespNET: A dual-route exploratory CNN model for pediatric adventitious respiratory sound identification

Pediatric respiratory diseases significantly contribute to the global burden of morbidity and mortality among children. Moreover, the long-term persistence of respiratory diseases from childhood into adulthood underscores the critical importance of early identification and diagnosis of such diseases. Auscultation, the widely recognized diagnostic method for respiratory diseases, relies on expertise and is subject to variability across different practitioners. To address these challenges, extensive research has focused on automating respiratory disease detection; however, the majority of existing studies have focused on adult populations. Therefore, in this study, we utilize the SPRSound database, which comprises respiratory events from 288 youths aged between 1 month and 18 years to develop deep learning-based models for automatic adventitious lung sound detection in children. Initially, we explore various architectures, including convolutional neural networks with and without transformer encoders, as well as vision transformer. Building upon these investigations, we propose a novel spectrotemporal deep neural network called TRespNET, which incorporates both Mel-scale spectrograms and raw time series of sound recordings. In addition, we integrate hand-crafted acoustic features with the proposed neural network feature maps for further model development. Our results yield a specificity of 0.98, a sensitivity of 0.84, and a harmonic score of 0.90, demonstrating the superiority of the proposed method over all existing methods on the SPRSound dataset. This research provides remarkable insights into an automated and accurate approach for respiratory disease classification in children.

Acoustic Echo Cancellation using Adaptive Filter for Quranic Accent Signals

Quranic recordings andechoed portions of the emphasis are susceptible to signal reverberation, particularly when being listened to in a conference room. Tajweed and Quranic verse rule identification are susceptible to additive noise, which could lower classification accuracy. In order to reflect the most correct rate following pattern categorization, this study suggested the appropriate use of three adaptive algorithms: Affine Projection (AP), Least Mean Square (LMS), and Recursive Least Squares (RLS). For feature extraction, Mel Frequency Cepstral Coefficient is used together with Probabilities Principal Component Analysis (PPCA), K-Neural Network (KNN) and Gaussian Mixture Model (GMM). AP indicates 93.9% for all of the classification algorithm in used, while for LMS and RLS the results are differed varies on different pattern classification algorithm stated whereby with LMS and PPCA classification, 96.9 % for accuracy and 84.8% accuracy for LMS and KNN. While for RLS and GMM, 96.9% was achieved and the results were reduced for both KNN and PPCA. The analysis has resulted for both on accuracies within different filtering algirithm and classification for accuracy and ERLE(dB).Towards this research it is hope will embark more understanding towards echo cancellation and quality of sound recordings that may affected even to the Quranic recordings.KeywordsAdaptive filtering, acoustic echo cancellation, recursive least squares, least mean square, affine projection, accuracy rate1.IntroductionUnusually big peaks have been produced by frequency response, masking, and emerging peaks from sound systems. Noise can be introduced into music recordings from the surrounding environment, during the recording system when audio signals change, or during the indexing process. These noise signals can disrupt and lower the quality and efficiency of the recording. Quranic recordings may also be affected by echoes, which can interfere with the recording process. The speaker, microphone, and transmission path are among the variables that affect.

Interrogating "Access to Waxes" – Introductory Remarks on the Collections from the Arab World of the Berlin Phonogramm-Archiv

This paper contextualizes the “Access to Waxes” workshop and publications within the broader context of the ongoing debates in ethnomusicology and related disciplines about the ethics and politics of historical sound recordings, archives, acoustic memory, cultural rights, intellectual property, musical repatriation, and access. Selected initiatives on the topic spearheaded by the two major scholarly societies focusing on the transdisciplinary study of music, the ICTMD and the Society for Ethnomusicology, are referred to. The contribution also critically evaluates the categorization and labeling of music in archives (like the term “Arab/Arabic music”) and emphasizes the need for a critical study of historical sound recordings like the collections of the Berlin Phonogrammarchiv. This would involve collaboration with source communities and local institutions ideally leading to the development of shared research practices and creating the conditions to access, revitalize, and sustain their musical heritage.

Underwater vocalizations in foraging female Antarctic fur seals (Arctocephalus gazella) in the Kerguelen Islands

AbstractIn the marine environment, many animals use sound to interact and communicate with their conspecifics or other species. Over the last decades, the development of sound recording systems has allowed a significant advance in our knowledge of sound production in marine animals. We deployed miniature sound and movement tags on Antarctic fur seals (AFS), a small otariid foraging on mesopelagic fish, to uncover potential underwater vocalizations in this species. Tags recorded underwater sounds synchronously with high‐resolution AFS movements and diving behavior, allowing us to investigate the behavioral context of vocalizations in the natural environment. We provide evidence of underwater vocalizations in foraging female AFS in a context of foraging at sea. AFS produced stereotyped calls made of low‐frequency pulses produced in series, exclusively during foraging dives. We hypothesized that these acoustic pulse series could be used as an acoustic lure to confuse or attract fish prey, however, a larger sample size is needed to study the adaptive significance of these underwater vocalizations.

Analysis of unsupervised learning approach for classification of vehicle fuel type using psychoacoustic features

Much research has been done in the field of classifying vehicles based on their fuel type. One of the many potential applications is to improve the quality of life in urban areas by separating vehicles based on their pollution level. Real-time classification and implementation of appropriate on-site IoT measurement devices is critical to developing a system that accurately identifies vehicle's fuel type without violating driver privacy. In this paper, a classification system based on psychoacoustic features extracted from sound recordings is investigated. Unsupervised learning was implemented as it is able to detect hidden connections within the input space without relying on labelling data. Our goal was to develop and explore a relatively fast classification system, focusing on a short acquisition time. A self-organizing map with a 10-dimensional input space was used and effective classification with five single-valued features was demonstrated. The analysis of the input space shows the difficulty and complexity of such an approach and leaves room for further improvement.