Wave Files Research Articles

Defined radio wave frequencies attenuate the head-twitch response in mice elicited by (±)-2,5-dimethoxy-4-iodoamphetamine.

Results from clinical trials show that serotonergic psychedelics have efficacy in treating psychiatric disorders, where currently approved pharmacotherapies are inadequate. Developing psychedelic medicines, however, comes with unique challenges, such as tempering heightened anxiety associated with the psychedelic experience. We conceived a new strategy to potentially mitigate psychedelic effects with defined electromagnetic signals (ES). We recorded the electromagnetic fields emitted by the serotonin 2 receptor (5-HT2R) agonist (±)-2,5-dimethoxy-4-iodoamphetamine (DOI) and converted them to a playable WAV file. We then exposed the DOI WAV ES to mice to assess its effects on the DOI-elicited, 5-HT2AR dependent head-twitch response (HTR). The DOI WAV signal significantly attenuated the HTR in mice elicited by 0.1 and 0.3 mg/kg subcutaneous DOI (p < 0.05 and p < 0.01, respectively). A scrambled WAV signal did not affect the DOI-elicited HTR, suggesting specificity of the DOI WAV signal. These results provide evidence that defined ES could modulate the psychoactive effects of serotonergic psychedelics. We discuss putative explanations for the distinct effects of the DOI WAV signal in the context of previous studies that demonstrate ES's efficacy for treating other conditions, including pain and cancer.

A sound recognition algorithm to filter non construction site noise

Sixense Engineering have developed a new algorithm to correct the noise levels measured around construction sites, by identifying and filtering out sounds that are generated by activities outside the site (emergency services sirens, traffic horns, etc.). The idea is to go beyond the simple classification of audio files by applying the classification algorithm to all the noise events that contribute to the noise dose LAeq,T. In this innovative process, the noise dose is calculated from elementary LAeq,1min,. When noise events are detected (in terms of LAeq,1min), the corresponding audio recording (.wav file) are analysed by the sound recognition algorithm. When the algorithm identifies that the noise source is not construction related, with a high degree of confidence, the corresponding LAeq,1min is excluded from the noise dose calculation. In other words, the measured noise dose can be filtered by removing non construction related noise events produced by external sources like sirens, horns, motorbikes, birds, music, etc.

Characterization of maritime traffic noise in two areas of the Mediterranean

One of the objectives of the European JPI Oceans DeuteroNoise project is to investigate the soundscapes within different sea basins. Measurement campaigns, including the recording of raw noise data (wave files), have been carried out in the Venice Lagoon, in the North Adriatic and on Barcelona shore. In this work we will present the first results. The data acquired were processed to obtain the characteristic spectrum and noise levels; subsequently the velocity of the particles will be acquired and analysed. The analysis of the recorded data involves the design of a taxonomy of anthropogenic noise sources, e.g. large ships, small boats or any other existing noise source. The recorded data were labelled with these sources, describing their main audio characteristics (spectral distribution, temporal evolution, etc.) for their characterization and future blind identification. Throughout the measurement period, ship and route data were acquired using AIS (Automatic Identification System) and Vessel Monitoring System (VMS) in the measurement sites. The preliminary analysis and labelling, together with ship and route data, will allow the reproduction of acoustic environments of anthropogenic origin in laboratories and the creation of acoustic maps of the basins investigated using commercial software.

Road-pavement classification by artificial neural network model based on tire-pavement noise and road-surface image

This study focuses on an artificial neural network (ANN) model for classifying pavement types using acoustic and image data. While conventional studies often use road-surface images for pavement classification, they face challenges with image quality degradation owing to external factors, such as sunlight angle, shadows, and lighting. Therefore, in this study, tire-pavement noise, which has different noise characteristics depending on the material and surface treatment, is used independently and in conjunction with image data for ANN training. To construct the training dataset, tire-pavement noise, and road-surface images are collected from 11 highway sampling sites in South Korea. Two simultaneous measurements are used: the tire-pavement noise is collected using the On-board sound intensity (OBSI) method, and the camera captures the road-surface images. 1/3 octave SIL, spectrum, MFCC, GLCM, and HOG are extracted from the raw data, and the ANN models are trained by these features. Using the spectrum as an input feature for the ANN yields a classification accuracy of 95.18%. However, the total number of parameters in the ANN is double that of the other models. To reduce the ANN size, 1/3 octave band SIL is used for training, and the model size is halved. However, the accuracy decreases by 13.47 percentage points. To overcome this significant decrease, the 1/3 octave bands SIL and image features were used to train ANN, simultaneously. This approach increases the accuracy by 93.85%. By training the ANN using MFCC, which is commonly used as an acoustic feature in other machine learning studies, the highest classification accuracy of 96.84% is achieved. Additionally, MFCC models are affected by the number of coefficients and the signal length. To include the dominant frequency of tire-pavement noise, more than 13 coefficients are used, a number generally known to be suitable for speech recognition. Increasing the number of coefficients from 13 to 40 improves accuracy by 1.17 percentage points. The interval for slicing raw WAV files is reduced to increase the training data and classify the pavement using shorter signals without statistically significant accuracy loss. Although accuracy does not decrease until the signal lengths reach 0.5 seconds, it rapidly decreases when the signal lengths become shorter than 0.4 seconds.

Assessment of Self-Supervised Denoising Methods for Esophageal Speech Enhancement

Esophageal speech (ES) is a pathological voice that is often difficult to understand. Moreover, acquiring recordings of a patient’s voice before a laryngectomy proves challenging, thereby complicating enhancing this kind of voice. That is why most supervised methods used to enhance ES are based on voice conversion, which uses healthy speaker targets, things that may not preserve the speaker’s identity. Otherwise, unsupervised methods for ES are mostly based on traditional filters, which cannot alone beat this kind of noise, making the denoising process difficult. Also, these methods are known for producing musical artifacts. To address these issues, a self-supervised method based on the Only-Noisy-Training (ONT) model was applied, consisting of denoising a signal without needing a clean target. Four experiments were conducted using Deep Complex UNET (DCUNET) and Deep Complex UNET with Complex Two-Stage Transformer Module (DCUNET-cTSTM) for assessment. Both of these models are based on the ONT approach. Also, for comparison purposes and to calculate the evaluation metrics, the pre-trained VoiceFixer model was used to restore the clean wave files of esophageal speech. Even with the fact that ONT-based methods work better with noisy wave files, the results have proven that ES can be denoised without the need for clean targets, and hence, the speaker’s identity is retained.

Biases in Ecoacoustics Analysis: A Protocol to Equalize Audio Recorders.

Eco-acoustic indices allow us to rapidly evaluate habitats and ecosystems and derive information about anthropophonic impacts. However, it is proven that indices' values and trends are not comparable between studies. These incongruences may be caused by the availability on the market of recorders with different characteristics and costs. Thus, there is a need to reduce these biases and incongruences to ensure an accurate analysis and comparison between soundscape ecology studies and habitat assessments. In this study, we propose and validate an audio recording equalization protocol to reduce eco-acoustic indices' biases, by testing three soundscape recorder models: Song Meter Micro, Soundscape Explorer Terrestrial and Audiomoth. The equalization process aligns the signal amplitude and frequency response of the soundscape recorders to those of a type 1 level meter. The adjustment was made in MATLAB R2023a using a filter curve generated comparing a reference signal (white noise); the measurements were performed in an anechoic chamber using 11 audio sensors and a type 1 sound level meter (able to produce a .WAV file). The statistical validation of the procedure was performed on recordings obtained in an urban and Regional Park (Italy) assessing a significant reduction in indices' biases on the Song Meter Micro and Audiomoth.

DCNN for Pig Vocalization and Non-Vocalization Classification: Evaluate Model Robustness with New Data.

Since pig vocalization is an important indicator of monitoring pig conditions, pig vocalization detection and recognition using deep learning play a crucial role in the management and welfare of modern pig livestock farming. However, collecting pig sound data for deep learning model training takes time and effort. Acknowledging the challenges of collecting pig sound data for model training, this study introduces a deep convolutional neural network (DCNN) architecture for pig vocalization and non-vocalization classification with a real pig farm dataset. Various audio feature extraction methods were evaluated individually to compare the performance differences, including Mel-frequency cepstral coefficients (MFCC), Mel-spectrogram, Chroma, and Tonnetz. This study proposes a novel feature extraction method called Mixed-MMCT to improve the classification accuracy by integrating MFCC, Mel-spectrogram, Chroma, and Tonnetz features. These feature extraction methods were applied to extract relevant features from the pig sound dataset for input into a deep learning network. For the experiment, three datasets were collected from three actual pig farms: Nias, Gimje, and Jeongeup. Each dataset consists of 4000 WAV files (2000 pig vocalization and 2000 pig non-vocalization) with a duration of three seconds. Various audio data augmentation techniques are utilized in the training set to improve the model performance and generalization, including pitch-shifting, time-shifting, time-stretching, and background-noising. In this study, the performance of the predictive deep learning model was assessed using the k-fold cross-validation (k = 5) technique on each dataset. By conducting rigorous experiments, Mixed-MMCT showed superior accuracy on Nias, Gimje, and Jeongeup, with rates of 99.50%, 99.56%, and 99.67%, respectively. Robustness experiments were performed to prove the effectiveness of the model by using two farm datasets as a training set and a farm as a testing set. The average performance of the Mixed-MMCT in terms of accuracy, precision, recall, and F1-score reached rates of 95.67%, 96.25%, 95.68%, and 95.96%, respectively. All results demonstrate that the proposed Mixed-MMCT feature extraction method outperforms other methods regarding pig vocalization and non-vocalization classification in real pig livestock farming.

Proposed CNN Model for Audio Recognition on Embedded Device

The audio detection system enables autonomous cars to recognize their surroundings based on the noise produced by moving vehicles. This paper proposes the utilization of a machine learning model based on convolutional neural networks (CNN) integrated into an embedded system supported by a microphone. The system includes a specialized microphone and a main processor. The microphone enables the transmission of an accurate analog signal to the main processor, which then analyzes the recorded signal and provides a prediction in return. While designing an adequate hardware system is a crucial task that directly impacts the predictive capability of the system, it is equally imperative to train a CNN model with high accuracy. To achieve this goal, a dataset containing over 3000 up-to-5-second WAV files for four classes was obtained from open-source research. The dataset is then divided into training, validation, and testing sets. The training data is converted into images using the spectrogram technique before training the CNN. Finally, the generated model is tested on the testing segment, resulting in a model accuracy of 77.54%.

Accelerating dynamic time warping for speech recognition with SSE

This study presents a significant enhancement to the Dynamic Time Warping (DTW) algorithm for real-time applications like speech recognition. Through integration of SIMD (Single Instruction Multiple Data) instructions to distance function, the research demonstrates how SSE accelerates DTW, markedly reducing computation time. The paper not only explores the theoretical aspects of DTW and this optimization but also provides empirical evidence of its effectiveness. Diverse dataset of 18 voice command classes was assembled, recorded in controlled settings to ensure audio quality. The audio signal of each speech sample was segmented into frames for detailed analysis of temporal dynamics. DTW search was performed on features set based on Mel Frequency Cepstral Coefficients (MFCC) and Linear Predictive Coding (LPC), combined with delta features. A comprehensive set of 27 features was extracted from each frame to capture critical speech characteristics. The core of the study involved applying traditional DTW as a baseline for performance comparison with the SSE-optimized DTW. The evaluation, focusing on computational time, included measurements like minimum, maximum, average, and total computation times for both standard and SSE-optimized implementations. Experimental results, conducted on datasets ranging from 5 to 60 WAV files per class, revealed that the SSE-optimized DTW significantly outperformed the standard implementation across all dataset sizes. Particularly noteworthy was the consistent speed of the SSE-optimized Manhattan and Euclidean distance functions, which is crucial for real-time applications. The SSE-optimized DTW maintained a low average time, demonstrating remarkable stability and efficiency, especially with larger datasets. The study illustrates the potential of SSE optimizations in speech recognition, emphasizing the SSE-optimized DTW's capability to efficiently process large datasets.

A Novel Hybrid Method for Effective Identification and Extraction of Digital Evidence Masked by Steganographic Techniques in WAV and MP3 Files

Anti-forensics techniques, particularly steganography and cryptography, have become increasingly pressing issues affecting current digital forensics practices. This paper advances the automation of hidden evidence extraction in audio files by proposing a novel multi-approach method. This method facilitates the correlation between unprocessed artefacts, indexed and live forensics analysis, and traditional steganographic and cryptographic detection techniques. In this work, we opted for experimental research methodology in the form of a quantitative analysis of the efficiency of the proposed automation in detecting and extracting hidden artefacts in WAV and MP3 audio files. This comparison is made against standard industry systems. This work advances the current automation in extracting evidence hidden by cryptographic and steganographic techniques during forensic investigations. The proposed multi-approach demonstrates a clear enhancement in terms of coverage and accuracy, notably on large audio files (MP3 and WAV), where manual forensic analysis is complex, time-consuming and requires significant expertise. Nonetheless, the proposed multi-approach automation may occasionally produce false positives (detecting steganography where none exists) or false negatives (failing to detect steganography that is present). However, it strikes a good balance between efficiently and effectively detecting hidden evidence, minimising false negatives and validating its reliability.

Collaboration with local fieldworkers to support remote collection of high quality audio speech data

In 2022 we planned speech data collection with speakers of Syrian and Jordanian dialects to inform an updated Syrian Arabic dialectology in response to sustained displacement of millions of Syrians. The pandemic imposed remote data collection, but an internet-based approach also facilitated recruitment with this highly distributed speech community. Their vulnerable situation brings barriers, however, since most prospective participants have limited internet data and rarely use email. We collected self-recorded short audio files in which participants read scripted materials and described pictures. Three platforms were tested: Gorilla, Phonic and Awesome Voice Recorder (AVR, smartphone app). Gorilla/Phonic offer stimulus presentation advantages, so were piloted thoroughly, but the audio quality obtained was not suitable for phonetic analysis, ruling out their use in the main study. AVR yields full spectrum wav files but requires participants to submit files by email, so we recruited local fieldworkers to support participants with recording and file submission. We asked fieldworkers and participants about their experience of working with us, through surveys and interviews. The results confirm fieldworker involvement was crucial to the success of the project which generated high quality audio data, suitable for phonetic analysis, from 134 speakers within three months (Almbark, Hellmuth, &amp;amp; Brown, forthcoming).

A WAV file dataset of bottlenose dolphin whistles, clicks, and pulse sounds during trawling interactions

Globally, interactions between fishing activities and dolphins are cause for concern due to their negative effects on both mammals and fishermen. The recording of acoustic emissions could aid in detecting the presence of dolphins in close proximity to fishing gear, elucidating their behavior, and guiding potential management measures designed to limit this harmful phenomenon. This data descriptor presents a dataset of acoustic recordings (WAV files) collected during interactions between common bottlenose dolphins (Tursiops truncatus) and fishing activities in the Adriatic Sea. This dataset is distinguished by the high complexity of its repertoire, which includes various different typologies of dolphin emission. Specifically, a group of free-ranging dolphins was found to emit frequency-modulated whistles, echolocation clicks, and burst pulse signals, including feeding buzzes. An analysis of signal quality based on the signal-to-noise ratio was conducted to validate the dataset. The signal digital files and corresponding features make this dataset suitable for studying dolphin behavior in order to gain a deeper understanding of their communication and interaction with fishing gear (trawl).

“TimBre” Pilot Study Conducted Using Multi-Country Training and Validation Data for Screening of Pulmonary Tuberculosis Using Cough (Acoustic Sounds), Clinical &amp; Demographic Inputs

TimBre from Docturnal offers screening for multiple lung diseases – Pulmonary Tuberculosis, Pneumonia, Covid19 &amp; COPD. Detailed studies of TimBre in the past used a third-party Microphone Array that focused on a XY arrangement that provided high fidelity cough sounds with an average length of &gt;5 seconds and demographic data such as Height, Weight, BMI [1]. In the current study, cough sounds were collected from 7 different countries (India, Vietnam, Philippines, Uganda, Tanzania, Madagascar, and South Africa) using Mobile Phones from different manufacturers &amp; recorded solicited coughs in a clinic for a duration of 0.5 seconds. A plethora of demographic and clinical variables were provided of which a subset was used by TimBre algorithm. Most importantly, the .WAV files were recorded in a single channel at a sampling rate of 44.1kHz &amp; 16 bits. The study details two approaches wherein the first method was to concatenate all the 0.5 second WAV files based on a timestamp provided for each StudyID in the training &amp; scoring set while the second method involved using the 0.5 second snippets as-is in both training and validation sets without any concatenation. Both approaches used a combination of demographic, clinical and spectral variables. The first approach on the independent test set yielded a sensitivity and specificity of 68.6% and 71.7% respectively with an AUC of 0.75 while the second approach yielded a sensitivity &amp; specificity of 75.41% and 68.30% respectively with an AUC of 0.78. Thus, the ML model performed better in the second approach and we anticipate it to improve with additional training data.

Detecting Emotions in Audio Data of Patients with Post Traumatic Stress Disorder using Convolutional Neural Networks

As humans, we have an effortless ability and a high accuracy to identify another human's emotions through the tone, pitch and pace of their speech, even the emphasis and stress placed on each word. However, people with a traumatic experience suffering from Post Traumatic Stress Disorder (PTSD), 24.4 million people in the USA, can often repress emotions making it difficult for therapists to identify their patients genuine emotions and to treat them appropriately. Using an upcoming field of emotion detection in Artificial Intelligence (A.I.), I identified the human emotions from speech. Instead of using an audio transcription based model, I opted for a newer image based model, RESNET-18, which is widely used and utilizes spectrograms to preserve the subtleties in speech, critical in distinguishing emotions. To train the model, I used the RAVEDESS dataset which consists of wav files with eight different emotions. I was able to achieve an overall accuracy of 82% (greater than human detection by 25%). Specifically, I achieved 99% for no stress class (happiness), 97% for neutral class, (neutral, calm, and surprised), and 8\5% for stressed class (fearful, sadness, anger, disgust). I also found that the model got an accuracy of 87% when only trained on males, with continued training an overall accuracy above 90% is definitely achievable. In conclusion, it is possible to find the emotions of PTSD patients, and in the future, continued research can help improve the lives of people who are not able to express their true emotions.

A Smart Control System for the Oil Industry Using Text-to-Speech Synthesis Based on IIoT

Oil refineries have high operating expenses and are often exposed to increased asset integrity risks and functional failure. Real-time monitoring of their operations has always been critical to ensuring safety and efficiency. We proposed a novel Industrial Internet of Things (IIoT) design that employs a text-to-speech synthesizer (TTS) based on neural networks to build an intelligent extension control system. We enhanced a TTS model to achieve high inference speed by employing HiFi-GAN V3 vocoder in the acoustic model FastSpeech 2. We experimented with our system on a low resources-embedded system in a real-time environment. Moreover, we customized the TTS model to generate two target speakers (female and male) using a small dataset. We performed an ablation analysis by conducting experiments to evaluate the performance of our design (IoT connectivity, memory usage, inference speed, and output speech quality). The results demonstrated that our system Real-Time Factor (RTF) is 6.4 (without deploying the cache mechanism, which is a technique to call the previously synthesized speech sentences in our system memory). Using the cache mechanism, our proposed model successfully runs on a low-resource computational device with real-time speed (RTF equals 0.16, 0.19, and 0.29 when the memory has 250, 500, and 1000 WAV files, respectively). Additionally, applying the cache mechanism has reduced memory usage percentage from 16.3% (for synthesizing a sentence of ten seconds) to 6.3%. Furthermore, according to the objective speech quality evaluation, our TTS model is superior to the baseline TTS model.

Enhancing Pipe-Break Early Warning in Smart Water Networks: Distinguishing Leaks from Water Uses

This research presents a denoising technique developed for enhancing the identification of newly developed and/or developing leaks by acoustic loggers in smart water networks. The key challenge addressed is the differentiation of leak-induced signals from signals originating from other sources, such as customer water use, pumps operations and environmental noise. A spectral subtraction-based denoising technique is adapted to process the acoustic waves measured daily using wireless accelerometers. A newly captured wave file can be filtered based on a reference wave file, either one with a known nonleak noise source or one measured in the past at the same location, to highlight the differences or the evolution of the signals over time. This technique enhances the robustness of automated alarms in identifying leaks in water networks.

KHiTE: Multilingual Speech Acquisition to Monolingual Text Translation

Objectives: To develop a system that accepts cross-lingual spoken reviews consisting of two to four languages, translate to target language text for Indic languages namely Kannada, Hindi, Telugu and/or English termed as cross lingual speech identification and text translation system. Methods: Hybridization of software engineering models are used in natural languages for pre-processing such as noise removal and speech splitting to obtain phonemes. Combinatorial models namely Hidden-Markov-Model, Artificial Neural Networks, Deep Neural Networks and Convulutional Neural Networks were deployed for direct and indirect speech mapping. Trained corpus consisting of thousand phonemes in the form of wave files for each language considered is named as KHiTEShabdanjali. The basic parameters cosidered for training dataset are pause, pitch, sampling frequency, threshold etc. Findings: The research has resulted in the development of mono-lingual and multi-lingual speech identification, tool for processing of cross-lingual speech and language identification, mono-lingual, bi-lingual, tri-lingual and quad-lingual speech to monolingual text translation for the four languages. It is a generic approach and can be used for other regional languages of India by training the corpus with the selected language. Novelty: Cross-lingual speech identification and text translation system helps users in e-shopping by reducing the time incurred in making a decision to purchase a product having enough features at an economical price, e-tutoring, e-farming activities, digitizing, defence etc. Keywords: Artificial Neural Networks; Convulutional Neural Networks; Deep Neural Networks; HiddenMarkovModel; Speech Processing

Using PRAAT in Teaching the Monophthongs to the EFL Students in Indonesia

&lt;p&gt;PRAAT can easily and rapidly improve the students’ ability in pronouncing monophthongs in phonetics class. The effectiveness of this software can help the students acquire quality monophthongs sounds. Furthermore, this software could reduce the problem of students learning sounds of monophthongs accurately such as the accuracy of F1 and F2 that is contained in monophthongs. It provided many programs on how to record, analyze, measure, and save in the wave file. This research aimed to investigate to what extent this software affects teaching monophthongs sounds in phonetics class for students. This research was conducted with 40 samples of students. They divided the class into 20 students for the control group and another 20 students for the experimental group. This quasi-experimental was to determine the best treatment before and after giving the treatment by using CALL for the experimental group and NON-CALL for the control group. Before using the software, the students were given the pre-test . Finally, they sat for post- test in the last meeting. The result was exposed through analysis of a t-test. It showed a significant difference in learning through PRAAT software (p&amp;lt;.05). In addition to that, teaching vowels sounds proved to be more successful using PRAAT than the traditional method.&lt;/p&gt;

Measuring and analyzing sound using phone microphone and free software

In Fall 2022 acoustics class, the students developed class projects using their smart phones to measure sound. In a laboratory exercise, the phones were calibrated against a standard microphone. A microphone circuit was bread boarded and calibrated to demonstrate the inner workings of electret microphone technology. Students developed projects around the measurement of sound, including measuring striking a cymbal, analyzing piano chords, analyzing guitar chords, characterizing reverberation time in a warehouse, and determining the amplification from a 3D printed human ear and bat ear. The sound files from the phone were edited to isolate the relevant sound and converted to WAV files using Garageband software. The WAV files were imported into Python for post-processing, such as windowing, filtering, and FFT. Most projects used a waterfall plot to demonstrate some time-frequency analysis. The ease with which the students mastered sound measurement and processing led to unexpectedly mature projects. The low cost of the equipment makes this approach a widely deployable educational paradigm.

Educational high school physics demonstration apparatus for acoustic landmine simulant resonance detection

This landmine “simulant” resonance detection demo was created in conjunction with a high school physics mentorship, using a variety of easily-accessibleequipment in order to demonstrate an important application of fundamental concepts of sound and vibration. A plastic tray (4 in. deep) contained a 3-in. diameter drum-like circular landmine simulant (1.5 in. high) with a thin acrylic top plate buried (1 in. deep) in a 3 in. layer of dry sifted masonry sand. A small accelerometer was place on the surface over the simulant. Airborne sound, a loud speaker chirp, with frequency slowly increasing over time, was created using a Mathematica® generated wav file played by the Raspberry Pi® computer. Sound excited the sand vibration that coupled to the top plate. The accelerometer generated a frequency response voltage, related to the sweep time. The Raspberry Pi® also played a role in implementing the A to D conversion, so that the frequency response near resonance was displayed as a voltage in a readable format. With this demo apparatus high school students will be excited by the fundamentals of wave propagation and concepts about resonance with regards to its real-life applications. [Specially inspired by James M. Sabatier’s research on acoustic landmine detection.]