Audible Signal Research Articles

Normative values and prediction equations for the modified incremental step test in healthy adults aged 18–83 years

ObjectiveTo establish normative values and reference equations for predicting the number of steps and oxygen consumption (VO2) from the modified incremental step test (MIST) in healthy adults aged 18–83 years. DesignProspective observational study. ParticipantsOne hundred and ninety-four healthy adults aged 18–83 years with normal spirometry. SettingsExercise physiology laboratory of a university. MethodsParticipants underwent two MISTs (30 minutes apart). The MIST was performed on a 20-cm-high step using an externally paced rhythm imposed by audible signals, starting with 10 steps per minute and with constant increments of 1 step every 30 seconds. Main outcomesNumber of steps and VO2 obtained from MIST. ResultsNormative values were provided for males and females for each age group. The following equations were determined: number of steps = 675.113 + (66.165*sex, 0 female and 1 male) - (5.353*age) - (6.593*body mass index) (R² =0.44, P < 0.001); VO2 = 0.106 + (0.216*sex, 0 female and 1 male) - (0.008*age [years]) + (0.021*weight [kilograms]) + (0.001*number of steps) (R² =0.80, P < 0.001). ConclusionNormative values and prediction equations are proposed for the number of steps and VO2 which can be used to interpret performance on the MIST in individuals with different health conditions. These equations now need validation in other samples.

MODEL OF PEDESTRIAN CROSSING BEHAVIOUR BASED ON ROAD TRAFFIC AND HUMAN FACTORS: A CASE STUDY OF MALAYSIA’S SHAH ALAM CITY

The present study intends to develop pedestrian crossing behaviour models based on road traffic and human factors. A questionnaire was distributed to 663 pedestrians in Shah Alam. Respondents were asked to complete a questionnaire regarding their risk perceptions and attitudes pertaining to walking and road crossings. This study identified two human factors that influenced pedestrian crossing behaviour: the “risk-taker” and the “rule-follower.” The modelling analysis revealed a substantial correlation between human factors and crossing behaviour. Analysis of pedestrian crossing behaviour is useful to evaluate the implementation of novel pedestrian crossing environments. The study offers insights applicable to urban planning and policy approaches for reducing pedestrian accidents by utilising strategies such as extended signal timing, audible signals, countdown timers, and optimising intersection design to improve pedestrian safety.

Design and Fabrication of a Smart Monitoring System for Liquefied Petroleum Gas-Operated Cars Based on Global System Mobile

In this paper, a low-cost smart system is designed and fabricated to monitor gas leakage in LPG-based cars. The implemented system consists of three gas sensors installed in the joints of the gas pipes, gas cylinders, and injectors for the purpose of gas leak monitoring. In addition, the LM 35 temperature sensor is situated over the gas cylinder to monitor its temperature. The implemented system can provide three types of alarm: the first type sends a text message to the car owner’s phone and emits an audible alarm signal. The message reaches the service center also. The second type sends messages via Bluetooth to the user’s phone or to the smart screen on the vehicle’s dashboard, with an alert that the engine must be turned off (emergency case). The last warning sends an overheating alarm signal, utilizing a temperature sensor when the temperature of the gas cylinder exceeds 50°C to provide protection and prevent a possible explosion. In practice, it was found that gas sensors should be placed less than 5 cm from the monitored site, and the best performance at work was under 55°C. It was also found that the best performance of the GSM unit is under 55°C.

Equine Gunshot Euthanasia: Creation of a 3D-Printed Model with Integrated Sensors for Training.

Challenges and issues related to the use of pentobarbital euthanasia and disposal of animal remains within the US have recently been reviewed. Environmental and public health challenges increasingly necessitate consideration of alternative methods such as gunshots, an American Veterinary Medical Association (AVMA) "acceptable with conditions" method, for the humane euthanasia of horses. A recent study reported a correctly aimed gunshot provides a humane option for euthanizing horses. However, although aiming guidelines exist, studies examining bullet trajectories in animals euthanized by gunshot have reported that inadequate disruption of the brain is a serious welfare issue. Here, we report the development and production of a portable, reusable, equine gunshot euthanasia training model. Using 3D printing, an anatomically accurate model of an equine head has been developed, with external aiming landmarks and equipped with integrated laser sensors and LED eyes. The laser sensors are embedded in two specific anatomical tracts (pons and medulla) with aiming paths associated with the aiming landmarks to train correct aiming angle. The LED eyes are linked to the laser sensors to provide instant feedback on aiming accuracy. When a beam from a commercially available blue training gun laser travels along the correct aiming path and strikes the sensor inside the head, the lights in the model's eyes go out and there is an audible signal, providing immediate feedback on the accuracy of the shot. The model facilitates the training of veterinary personnel and first responders in successful gunshot euthanasia, providing instantaneous feedback on the likelihood of a shot causing immediate, humane death in a live animal.

Nondestructive characterization and artificial intelligence recognition of acoustic identifiers of ancient ceramics

Cultural heritage identity management is the most basic and important work in the process of cultural heritage protection. It is of great significance to provide a unique and identifiable digital identity for ancient ceramics. At present, the identification information of ancient ceramics is mainly composed of external visual characteristics, and there is no report on feature identification method that can reflect the properties of ancient ceramics. Audible sound signals not only have advantages in non-destructive testing, but also can be used as voiceprint information to identify, monitor and analyze ancient ceramics. In this paper, seven ancient ceramics and 12 similar modern ceramic cups are taken as research objects, and an acoustic identifier (AID) is constructed. We put forward a reliable acoustic identification method for ancient ceramics, and established a digital code of acoustic characteristics of ancient ceramics. The results show that audible sound waves can reflect the attribute information of ancient ceramics. Sufficient acoustic data combined with deep learning can not only accurately match the identity of ancient ceramics, but also detect the real-time identity information of ancient ceramics, and make a comparative analysis of its cracks and whether it has caused damage. This method can provide a variety of practical applications for audible signal feature recognition technology in the exhibition, protection, trading, recognition and safety management of ancient ceramics and other cultural relics.

Potential effects of audible sound signals including music on plants: A new trigger

Plants are highly sensitive organisms and can indeed benefit from specific sound signals in multi-layered processes. Scientific evidences have shown the potential applications of sound wave treatment in plant biology. However, there are some limitations to sound wave treatment that must be overcome. We still do not understand how do plants initially perceive and recognize sound signals, which is very critical to maximize the effectiveness of the use of sound treatment from practical viewpoint. Proper setup of sound treatment equipment and detailed understanding and evaluation of the effects of selected frequencies and intensities along with sound exposure times are also very crucial during sound treatment. More experimental studies with different models need to be done in a multidisciplinary approach toward establishing suitable mechanism for sound treatment application in agriculture production. The aim of this paper is to provide an overview of findings associated with potential effects of audible sound waves including music on different biological, physiological and biochemical processes in plants.

NEAR ULTRASONIC ATTACK AND DEFENSIVE COUNTERMEASURES

The practical implications of issuing inaudible voice commands. The research mapped each attack vector to a tactic or technique from the MITRE ATT&amp;CK matrix, covering enterprise, mobile, and Industrial Control System (ICS) frameworks. The experiment involved generating and surveying fifty near-ultrasonic audios to assess the attacks' effectiveness. Unprocessed commands achieved a 100% success rate, while processed commands achieved an 86% acknowledgment rate and a 58% overall executed (successful) rate. The research systematically stimulated previously unaddressed attack surfaces, aiming for comprehensive detection and attack design. Each ATT&amp;CK identifier was paired with a tested defensive method, providing attack and defense tactics. The research findings revealed that the attack method employed Single Upper Sideband Amplitude Modulation (SUSBAM) to generate near-ultrasonic audio from audible sources. By eliminating the lower sideband, the design achieved a 6 kHz minimum from 16-22 kHz while remaining inaudible after transformation. The research also investigated the one-to-many attack surface, exploring scenarios where a single device triggers multiple actions or devices. Furthermore, the study demonstrated the reversibility or demodulation of the inaudible signal, suggesting potential alerting methods and the possibility of embedding secret messages like audio steganography. A critical methodological advance included tapping into the postprocessed audio signal when the server demodulates the signal for comparison to both the audible and inaudible input signals to improve the actionable success rates.

The birth of noise in modern China: Radio, amateur engineering, and the sonic network

This article argues that noise was born in modern China with the advent of radio and radio broadcasting. The media specificity of radio required that any radio listener must be an acoustic engineer to some extent, dealing with “noise current,” “tunable hum,” and “interference” before tuning into the desired radio program. This novel experience of listening as engineering gave rise to a new conceptualization of noise not merely as unpleasant sounds (the music-noise dichotomy), but also in terms of modern information technology (the signal-noise dichotomy). Integrating both modes of understanding noise, this article further suggests that the affordable radio sets and the public-oriented radio broadcasting together constituted an unprecedented sonic network, one that transmitted heterogenous sonic information from all over the world back home and at the same time deprived of any individual’s agency in deciding what to listen to. Living in this sonic network, one’s body-mind had to function as if it were a radio set, exposed to overwhelming information streams and trying to filter out noise as both disturbing signals and undesired contents. Radio noise is therefore symptomatic of the most fundamental dilemma China was facing in an era of global modernity, national crisis, and information explosion.

Directional speakers as a tool for animal vocal communication studies.

Audio playbacks are a common experimental tool in vocal communication research. However, low directionality of sound makes it hard to control the audience exposed to the stimuli. Parametric speakers offer a solution for transmitting directional audible signals by using ultrasonic carrier waves. The targeted transmission of vocal signals offers exciting opportunities for testing the diffusion of information in animal groups and mechanisms for resolving informational ambiguities. We have field tested the quality and directionality of a commercial parametric speaker, Soundlazer SL-01. Additionally, we assessed its usability for performing playback experiments by comparing behavioural responses of free-ranging meerkats (Suricata suricatta) with calls transmitted from conventional and parametric speakers. Our results show that the tested parametric speaker is highly directional. However, the acoustic structure of meerkat calls was strongly affected and low frequencies were not reliably reproduced by the parametric speaker. The playback trials elicited weakened behavioural responses probably due to the partial distortion of the signal but also indicating the potential importance of social facilitation for initiating mobbing events in meerkats. We conclude that parametric speakers can be useful tools for directed transmission of animals calls but after a careful assessment of signal fidelity.

Increasing seat belt use in the United States by promoting and requiring more effective seat belt reminder systems

Objective Federal Motor Vehicle Safety Standard (FMVSS) 208 requires every passenger vehicle to provide an auditory signal lasting 4 to 8 seconds and a visual display lasting 60 seconds when the driver is unbelted at ignition. This requirement does not increase seat belt use. This paper summarizes the latest research on using vehicle technology to increase seat belt use and existing safety standards worldwide to support the strengthening of FMVSS 208. Method Studies of seat belt reminders and interlocks published in peer-reviewed journals, conference proceedings, or as technical reports were identified in online databases and reviewed along with current requirements worldwide. Results from past research were used to estimate the front- and rear-seat daytime belt use rate and the annual number of lives that could be saved by a persistent audible reminder at each seating position. Results Most motor vehicle occupants routinely buckle up. Those that do not typically forget, are going a short distance, or find belts uncomfortable. Seat belt reminders can remind or motivate occupants to buckle up. Enhanced reminders that exceed FMVSS 208 increase belt use by 6 percentage points. Reminders also can increase rear belt use, and although required throughout the world, are not required by FMVSS 208. More persistent reminders, like those required around the world, with a continuous, long-lasting audible signal increase belt use by 30% among drivers who do not routinely buckle up. If every vehicle in the U.S. had such a reminder at each seating position, then it was estimated that the daytime belt use rate in the U.S. would increase about 3 percentage points from 90.3% to 93.2% in the front row and by about 6 percentage points from 80% to 85.9% in the rear row. It was estimated that the increase in belt use from a continuous, long-lasting audible reminder could potentially save about 1,600 lives each year. Seat belt interlocks can increase belt use, but acceptance is a stumbling block. Public outcry ensued after interlocks were required in 1973, and public sentiment remains negative. Opinions toward front and rear reminders are more favorable. Furthermore, past research suggests interlocks may be no more effective for increasing seat belt use than persistent audible reminders. The effect of interlocks on rear belt use have not been explored. Conclusion Persistent seat belt reminder systems that last at least 90 seconds can potentially save hundreds of U.S. motorists each year. Robust empirical evidence, successful exemplars from organizations throughout the world, and a clear public health benefit exists for strengthening FMVSS 208 to require more persistent audible reminders at every seating position.

Evaluation of three pedestrian phasing with audible pedestrian signals configurations in Quebec City (Canada): an exploratory study of blind or visually impaired persons’ sense of safety, preferences, and expectations

Purpose This exploratory study aimed to evaluate the preferences, expectations, and sense of safety of blind or visually impaired persons regarding three types of pedestrian phasing with audible pedestrian signals configurations that exist in Quebec City (Canada). These include: 1) exclusive phasing with non-directional audible pedestrian signals; 2) exclusive phasing with directional audible pedestrian signals; and 3) concurrent phasing with directional audible pedestrian signals. Methods Thirty-two blind or visually impaired persons were asked to complete a survey. Their preferences and expectations regarding audible pedestrian signals were documented through a series of simulations. Their sense of safety regarding the three existing configurations were also documented. Subsequently, semi-directed, individual interviews with 11 of the individuals who had completed the survey were conducted to build off the collected information. Results No formal consensus regarding many of the issues discussed were established as participants’ responses varied too significantly. However, findings suggest that the exclusive phasing with directional audible pedestrian signals configuration is perceived to be the safest option by the participants. Conclusion This study may have practical implications on the design of intersections (e.g., selection of a type of pedestrian phasing with audible pedestrian signal) and the training of blind or visually impaired pedestrians. IMPLICATIONS FOR REHABILITATION The addition of audible pedestrian signals to pedestrian signals heightens the sense of safety of blind or visually impaired persons. This study may have practical implications on the design of intersections with audible pedestrian signals and the selection of a type of audible pedestrian signals based on intersection characteristics. Since many participants reported a lower sense of safety when faced with concurrent phasing, it is recommended that more intensive orientation and mobility interventions be provided to blind or visually impaired pedestrians who use this type of traffic signals. Educating road users about blind or visually impaired pedestrians appears also essential.

3D printed acoustic metamaterial filters for the mitigation of inaudible ultrasound attacks on smart speakers

Acoustic metamaterials have shown great potential in manipulating acoustic waves with a small footprint and versatile functionality. While their usefulness has been demonstrated in many fields including noise reduction and biomedical ultrasound, not many studies have been done in the field of audible acoustics and speakers. In this work, we show that an acoustic metamaterial-based filter can enhance the security of smart speakers, which are susceptible to inaudible ultrasound attacks due to the shadow effect of the microphones. The filter contains several resonant unit cells which collectively modulate the received signals in the ultrasound spectrum and, thus, mitigate inaudible attacks on smart speakers. On the other hand, normal audible signals are not affected and, therefore, regular functionality is not disturbed. The filtering effect is verified numerically using finite element analysis. Measurements are performed to validate the concept, where it is shown that inaudible ultrasound attacks are effectively blocked when the 3D-printed metamaterial filters are installed. This demonstration shows a possible way to apply acoustic metamaterials in consumer electronics such as smart speakers.

Extracting Spatial Muscle Activation Patterns in Facial and Neck Muscles for Silent Speech Recognition Using High-Density sEMG

Human is capable of producing versatile and precise speech owing to the complex neuromuscular systems that control the movement of facial and neck muscle groups. Silent speech recognition based on surface electromyography (sEMG) has been widely researched, which enables speech communication in the absence of an audible acoustic signal. However, the differences in the activities of facial and neck muscle groups under different phoneme pronunciation conditions have not been well investigated. The present work aims to quantify the macroscopic spatial patterns of the muscle activations at the phoneme level during the movement of facial and neck speech articulator muscle groups (SAMGs) using 320-channel high-density (HD) electrode arrays. 10 subjects performed 14 vowel speech tasks and 15 consonant speech tasks in audible speech and silent speech modes, respectively. Specifically, the root-mean-square heat maps of sEMG signals were used to characterize the SAMGs activation patterns. The results show that the muscle activation regions are overlapped under different phoneme pronunciation conditions and symmetric between the left and right sides of face and neck with the average correlation coefficient > 0.9. In addition, the best result can achieve a classification accuracy of 85.78% for consonant and 79.42% for vowel in silent speech mode. The overall results can provide a reference to the hardware or electrode design of portable device for practical use in future work.

Mitigating Inaudible Ultrasound Attacks on Voice Assistants With Acoustic Metamaterials

Voice assistants play an important role in facilitating human-machine interactions and have been widely used in audio consumer electronic products. However, it has been shown that they are susceptible to inaudible attacks in which the malicious signals are in the ultrasound regime and cannot be heard by human ears. In this study, we show that a judiciously designed acoustic metamaterial filter can mitigate such attacks by modulating the received signals by the microphones. The metamaterial filter is composed of rigid plates with individual holes which exhibit local resonance phenomena that suppress incoming waves at specific frequencies. The effectiveness of the metamaterial filter is confirmed by experiments which show that a combination of the holes can collectively distort the attack signals and protect the smart speakers. Moreover, normal audible signals are not affected by the proposed metamaterial, which adds to the flexibility of the device. The metamaterial filter has a small footprint and can be easily installed on various audio products. Our proposed strategy expands the capacity of acoustic metamaterials and improves the security of devices that use voice assistants.

Design of Wearable Patient Lying Position Tracking and Warning System to Prevent Pressure Injury

Within the scope of this study, a wearable lying position tracking system equipped with IMU sensors has been developed to prevent the formation of pressure injuries in bedridden patients. Three IMU sensors were placed on the patient's chest, one on the right upper leg and the other on the left upper leg, and the angular orientation expressions of the limbs were calculated. Datasets were created for three different hospitalization positions, and machine learning and deep neural network models were used to classify the patient's hospitalization type. The success of the classifiers was compared by calculating the accuracy, sensitivity, specificity, precision and F1 score values. The average accuracy values in the lying position classification were obtained as 99.506%, 99.977%, 99.972%, 99.838%, and 99.967% respectively, using Linear discriminant analysis, K-Nearest neighbor, Decision Tree, Support Vector Machine and Random Forest classification methods. The highest accuracy rate was obtained as a result of the K-Nearest neighbor method with high variation. The time that the person remained fixed in the determined lying position was also calculated, and if it remained longer than the specified time, an audible warning signal was generated to change the position. Thus, it has been tried to prevent the person to apply pressure by lying on a single muscle group and tissue for a long time and to prevent the formation of pressure injuries over time.

Virtual fencing of captive Asian elephants fitted with an aversive geofencing device to manage their movement

Aversive Geofencing Devices (AGDs) are designed to emit audible warning signals followed by electric shocks when animals reach virtual fences (VFs) with the intent that animals will learn to turn away at audio warnings and thereby avoid receiving shocks. AGDs are a potentially useful tool for mitigating human-elephant conflict, but a greater understanding of captive elephant responses to AGDs is required before they might be confidently used on wild elephants. We conducted experiments with eight, female captive Asian elephants using a modified dog-training collar to deliver mild electric shocks (4 kV) of varying strength (pulse frequencies) to determine the ideal location on the neck to deliver the stimuli and the optimum strength of the shock required to generate desired aversive responses. Ten shocks (<1 s duration) of different strengths were delivered during a 10 min session (i.e., one shock per minute) at two positions on one side of the elephant’s neck. Results indicated that elephants were more likely to display desirable aversive behaviours at the upper position tested on the neck (odds ratio=0.47, 95% CI 0.25–0.87, P = 0.018) and at higher stimuli strengths (odds ratio=1.03, 95% CI 1.01–1.04, P ≤ 0.001). A conditioning experiment was then conducted several months later with five of the same elephants. These were individually trained to walk along a ∼100 m path to a food reward on three consecutive days, wearing a dummy collar. On the next three days and on one other day few months later, the elephants were fitted with a similar shock collar (positioned at the upper neck location, and with the highest strength tested earlier) to determine if the AGD could prevent the elephants from accessing the food reward. Three VFs were established at ∼30 m, ∼50 m and ∼60 m points along the path. As the elephant approached the food, a mild audio warning, a more aggressive audio warning, and an electric shock was administered at the first, second and third VFs respectively. Warnings and shocks were not delivered if elephants heeded earlier warnings. A maximum of five such trials were attempted. The VFs successfully kept elephants from reaching the food 77.8% of the time, with elephants responding to the audio warnings and avoiding electric stimulation 47.2% of the trials. These findings suggest that AGDs are a promising method to manage elephant movement, but further research is needed to develop a reliable approach for wild elephants.

End-to-End Protein Normal Mode Frequency Predictions Using Language and Graph Models and Application to Sonification.

The prediction of mechanical and dynamical properties of proteins is an important frontier, especially given the greater availability of proteins structures. Here we report a series of models that provide end-to-end predictions of nanodynamical properties of proteins, focused on high-throughput normal mode predictions directly from the amino acid sequence. Using neural network models within the family of Natural Language Processing and graph-based methods, we offer atomistically based mechanistic predictions of key protein mechanical features. The models include an end-to-end long short-term memory (LSTM) model, an end-to-end transformer model, a graph-based transformer model, and an equivariant graph neural network. All four models show exceptional performance, with the graph-based transformer architecture offering the best results but at the cost of requiring a graph structure as input. Conversely, the LSTM and transformer models offer end-to-end sequence-to-property prediction capabilities, providing efficient avenues for protein engineering, analysis, and design. We compare our results against published data based on a Principal Neighborhood Aggregation graph neural network, revealing that the transformer model offers better performance while also being able to predict a large set of the first 64 normal mode frequencies, simultaneously. The use of the end-to-end transformer model may facilitate other downstream applications through the use of transfer learning, and it offers a comprehensive prediction of dynamical properties without any structural knowledge, directly from the amino acid sequence. We demonstrate a potential application in scientific sonification, where the normal mode frequencies are transposed to generate audible signals for a detailed analysis of subtle changes of protein sequences.

Leveraging audible and inaudible signals for pronunciation training by sensing articulation through a smartphone

• Under pronunciation principles, this paper presents a smartphone-based pronunciation training system for practicing monophthong regarding vowel articulation. • Speech and ultrasonic signals are captured and analyzed simultaneously to determine the tongue position and shape of the lips respecting uttering vowels. • The proposed system implementing a commercial smartphone is robust with high accuracy for practical scenarios. • The evaluation for its impact on user experience shows that the proposed system has raised learning engagement and enjoyment. Learning a foreign language pronunciation is the most challenging task for non-native speakers. Improving pronunciation based on feedback on pronunciation error scores is also not easy for learners. Our goal is to develop an alternative approach to pronunciation training that can point out articulation errors at the phoneme level to users through a smartphone-based system. Using self-correction as a means of ensuring learners' engagement, we focus on self-correction of pronunciation by adjusting articulation. In order to identify articulation concerning pronunciation, the system evaluates both audible and inaudible acoustic signals to examine fine-grained frequency-shifting direction of mouth movements and tongue position simultaneously. The result shows that the system can reach an average accuracy of 99.09% and is robust in different scenarios and genders. Additionally, the evaluation of the user study reveals that the proposed system provides positive user experiences and allows learners to improve their pronunciation more efficiently.

Atom Tones: investigating waveforms and spectra of atomic elements in an audible periodic chart using techniques found in music production

Atom Music was introduced in 2019 as a way to create unique audible tones for each atomic element that are direct translations of that element’s spectral lines. Each atomic element produces a unique spectral line pattern that can be recognized as the fingerprint of that element. Sonification is the process of translating non-audible data into audible signals as a way to gain an understanding of the original data. In this paper, sonification is applied to atomic spectra, using technology primarily from music production. These were applied to atomic spectra using additive synthesis methods and analyzed using digital audio workstations. Interest in the audible tones has primarily been in element identification through each tone, as well as those interested in musical interpretation. We investigate what insights can be made by observing the digital waveforms and spectra of each element tone. We consider whether there are patterns within the different element waveforms; if there is any correlation between the elements producing similar beat patterns; and if there are any harmonic relations between electron states that are represented by the spectra itself. Results indicate that this method could be a useful tool in investigating atomic structure.

Audible acoustic characteristics of epoxy resins with partial discharge

The epoxy resin is the main solid insulating material for electrical equipment. PD (partial discharge) is an important factor leading to epoxy resin aging or failure. Fault detection technology based on audible acoustic imaging has the advantages of real-time monitoring, visibility, and flexibility and is gradually being used in the field of electrical equipment insulation state monitoring. However, the audible voiceprint features of epoxy resins with typical PD defects are not clear, which is not conducive to effective detection and fault diagnosis. Therefore, in this study, the study of audible voiceprint features of epoxy resins with a metal burr, pollution, metal foreign matters, and surface discharge defects was carried out. The discharge audible signals were obtained, and their time and frequency domain features were analyzed. The results show that the discharge audible signals of four PD defects can reflect the occurrence of PD simultaneously with pulse current signals. But the characteristics are obviously different in the time domain, which are embodied in listening perception, persistence, amplitude, and phase. In the frequency domain, the frequency spectra of discharge audible signals of four defects are mainly 5–10 and 13–24 kHz. Furthermore, the three feature vectors of frequency spectrum complexity, high-frequency energy factor, and maximum frequency point are extracted, which can well distinguish the discharge audible signals of four PD defects.