Sound Quality Evaluation Research Articles

Objective quantification of sound sensory attributes in side-by-side vehicles using multiple linear regression models

The evaluation of sound quality is a pivotal area of research within audio and acoustics. The sound quality evaluation methods commonly used include both objective and subjective, the latter being time-consuming and costly as they rely on listening tests. This research work aims to investigate the use of predictive sound quality models as a way to objectively assess the Desire-to-buy of side-by-side vehicles, in a more efficient, faster, and less costly way than conventional methods. Multiple linear regression algorithms were used to validate the objective models derived from objective physical metrics and perceptual psycho-physical metrics. The sensory profile objective models reported in this paper were constructed using parsimonious linear Lasso and Elastic-net algorithms. Our results show that linear objective models effectively account for each of the perceptual attributes of the sensory profiles and the Desire-to-buy, while only requiring a few physical and psychophysical metrics.

Sound quality evaluation in special vehicles based on multiple linear regression

The issue of interior sound quality in specialized vehicles has received widespread attention, which could directly affect the riding comfort of the vehicle. This research primarily focuses on the objective quantification of subjective perception of sound qualitySQ in special vehicles. The noise of special vehicles under different working conditions is measured and analyzed, and psychoacoustic parameters and A-weighted sound pressure levels are calculated. The subjective evaluation of interior sound quality was conducted using the grading scale method. A multiple linear regression model is used to establish the correspondence between subjective perceptions and objective parameters. It provides a reference basis for evaluating the sound quality of special vehicles.

Sound quality evaluation and quantification for power seat operating noise

In recent years, vehicle interior condition at running has been quiet due to progress of noise reduction technology. On the other hand, quietness is not required only at running condition but also at parking where slight operating noise is generated by various devices. In this study, we then focused on an operating noise during power seat operation as one of the devices radiated noise in cabin. The impression to power seat operating noise were investigated through subjective evaluation test. For the test, power seat radiated noise at various operating modes and seat types were recorded binaurally and presented in the subjective evaluation test. Participants evaluated the presented sound quality in several evaluation terms such as "loudness," "pitch," and "discomfort." As the results, representative power seat operating mode, in which sound impression varies widely among presented sounds, was found out and suitable evaluation terms for describing the sound quality characteristics well were also extracted. In addition, the representative sound quality tendency was attempted to be evaluated quantitatively using psychoacoustic metrics from recorded sound pressure signal of the power seat.

Sound quality evaluation tendency according to driving situation

In near future, autonomous vehicles are expected to be more popular due to the technology development. Through this transition from active driving situation to passive driving, the drivers' psychological state may change. This may cause change of impression to vehicle interior sound. If the sound evaluation tendency is changed, we should improve autonomous vehicle interior sound by considering the difference. In this study, we investigated whether drivers' impression to interior sound change or not according to the driving situation through subjective evaluation test using a driving simulator. In the test, an original driving simulator was prepared to reproduce identical sounds at both active and passive driving situation. EV and ICEV interior sounds were presented in the simulator for the sound quality evaluation about discomfort and sporty. Participants drove the simulator under various driving conditions including acceleration and deceleration and evaluated the sound quality. As the result, interior sounds were found to be evaluated as more discomfort at passive driving situation than that at active situation. Furthermore, this tendency was observed regardless the driving experience of participants. Accordingly, more interior sound reduction will be necessary for satisfying the customer in the autonomous vehicle development.

Simulation of pulse width modulation strategies: validation based on sound quality evaluations

Though the noise emitted by an electrical machine is less loud than the one from a combustion engine, the existing literature shows it is not always more pleasant. This is because of its strong tonal content, due to deterministic excitations (e.g. electromagnetic slotting and switching effects, mechanical gear mesh effects). The main source of magnetic noise is the radiation of the stator submitted to electromagnetic forces, which are harmonic by nature. Their highest frequency components are generated by the Pulse Width Modulation (PWM) technique designed to vary the supply frequency and operating speed of the electric machine. Several PWM strategies exist, leading to specific harmonic spectra. A temporal simulation model has been developed to auralize the corresponding electromagnetic noise emitted by the machine, considering its mechanical and electrical parameters. A testbench has been adapted to test the relevance of the model, both from an electrical and acoustic point of view. The simulation model and the testbench can be used to generate virtual and real sound stimuli to be assessed in a jury testing. The subjective evaluations allow to assess the relevance of the auralization model, but also to identify the least annoying strategies.

Sound quality characteristics of reciprocating hermetic refrigeration compressors

In recent years, sound quality (SQ) has become an increasingly important factor in the home appliance industry, enhancing the customer experience and comfort. Manufacturers that prioritize SQ in their designs can gain a competitive advantage in the market and increase customer satisfaction. The compressor is one of the main sources of noise in a refrigerator. While manufacturers typically provide refrigerator sound quality and noise labels, research that specifically evaluates the compressor sound quality metrics at a component level is not common. Therefore, this study aims to isolate the compressor from the refrigerator to investigate SQ metrics for hermetic refrigeration compressors. To achieve this, compressors with different capacities are operated and recorded at different working speeds for different thermal conditions. Objective parameters are calculated and tested using time domain signals and analyzed with spectral analysis methods. In addition to the objective parameters, subjective results are obtained through jury tests consisting of 75 people of different genders and age ranges. By considering objective parameters of sound quality as input and sound quality evaluation values as output, a regression model is established. It is found that Loudness, Sharpness, and Degree of Modulation play an important role in compressor sound quality evaluations consider.

Prediction and analysis of interior sound quality of high-speed trains based on XGBoost

This paper primarily investigates the objective quantification of subjective perceptions of sound quality in high-speed trains. The study measures and analyzes the sound quality in high-speed trains under high-speed operating conditions and calculates psychoacoustic parameters. A subjective evaluation of the interior sound quality is conducted using the grading scale method. An Extreme Gradient Boosting (XGBoost) algorithm is utilized to construct a predictive model for sound quality in high-speed trains and get high predictive accuracy for sound quality evaluation. Employing AI-based evaluation models to evaluate the sound quality in high-speed trains offers a more effective method for studying the interior sound quality of high-speed trains.

Sound quality evaluation of small fan noise by tonal components

As Covod-19 has made online meetings from home offices the norm, the noise from small fans installed in information devices has increasingly become a factor disrupting a comfortable environment. In addition, as global warming accelerates, the use of new small fans called wearable fans has expanded, and the distance between small fans and humans has become even closer. In this study, we investigated whether T-TNR and T-PR, which are two methods for evaluating the tone characteristics of small fans, can be applied to the noise of projectors and wearable fans. We investigated a method for selecting dominant components and searched for a method for determining the harshness threshold using T-TNR and T-TPR.

"Tasting" sounds by A.I. Sommelier - Unsupervised machine learning of wine evaluation applied for sound quality evaluation

Wine produced countries in European Union categorize quality of wine as three classes by subjective judgements (called as "Appellation" in French) like taste, land of producer, type of grape seed and so on. In order to evaluate those wine quality, unsupervised machine learning with measured optical and chemical data of wine enables to categorize those 3 subjective classes at high accuracies. With those "A.I. Sommelier" techniques, vacuum cleaner sounds are categorized with psychoacoustic parameters and then modeling by image recognition AI of sFFT images correlate with annoyance rating at high rate. Although unsupervised machine learning technique is uncommon, it shows unsupervised machine learning technique is useful for sound quality evaluation

Interior sound quality evaluation and forecasting of passenger vehicles based on hybrid optimization neural networks

The interior noise of vehicles directly affects the comfort of the occupants, necessitating precise evaluation and control. Existing research has focused on constructing mappings between objective parameters and subjective perceptions of noise, where back propagation neural networks (BPNNs) are widely used due to their strong nonlinear mapping capabilities. However, the selection of initial weights and thresholds can affect the predictive accuracy of BPNN. This study developed a BPNN model optimized by an intelligent algorithm for predicting the level of subjective annoyance of passengers during the movement. Initially, objective parameters of interior noise were obtained through acoustic signal processing techniques, and five parameters were selected for studying subjective annoyance through correlation analysis and two-tailed tests. Meanwhile, the actual subjective ratings of passengers on interior noise were captured for subsequent training of the model and testing of the results. Finally, the established sparrow search algorithm (SSA) and genetic algorithm (GA) optimized BPNN were used to predict subjective evaluations. The predictive accuracy and efficiency of the model were significantly improved upon validation, providing a viable alternative to traditional passenger vehicle noise assessment experiments and valuable references for future noise control and optimization efforts. The experimental results are consistent with the view that the neural network model optimized with a mixture of intelligent algorithms is closer to the passenger’s subjective annoyance level having higher accuracy and efficiency.

Study on the sound quality of the electric vehicle powertrain under acceleration conditions

The evaluation and prediction of the sound quality (SQ) of electric vehicle (EV) powertrains are critical to the overall SQ of EVs. Firstly, a grouping method for noise samples is proposed to achieve rational grouping when SQ evaluations are performed using the grouped paired comparison method and its improvements. Secondly, aiming at the limitations of psychoacoustic parameters in predicting SQ under nonstationary conditions, a SQ prediction model based on the energy features of intrinsic mode functions (IMF) of signals is proposed. Finally, SQ evaluations are conducted, comparing the prediction performance of two SQ models based on energy features and psychoacoustic parameters. The prediction results show that the mean absolute percentage error (MAPE) of the model with energy features is 4.18%, while the MAPE of the model with psychoacoustic parameters is 8.88%, which demonstrates that energy features are superior in predicting the SQ of EV powertrains under acceleration conditions.

Objective sound quality evaluation of the electric two-stage centrifugal compressor for fuel cell vehicles-experimental investigations

Sound quality is one of the most challenging studies in the field of automotive NVH. The most important noise source in fuel cell vehicles (FCVs) is the air compressor, and its noise seriously affects the sound quality of FCVs. The centrifugal compressor is now the mainstream choice for commercial FCVs, but there are few studies on its sound quality evaluation using objective psychoacoustic parameters. Therefore, a comprehensive vibration-noise experiment is conducted on an electric two-stage centrifugal compressor (ETCC) for FCVs to evaluate the sound quality accurately, including all operating conditions. The sound quality is evaluated by four objective psychoacoustic parameters: loudness, sharpness, fluctuation strength and roughness, which are the most widely used indicators. The reasons for the variation of these acoustic parameters under dynamic operating conditions are accordingly explained. It can be concluded that the loudness and sharpness show a strong and positive correlation with the rotational speed, while the fluctuation strength and roughness just reflect the sound quality of the ETCC for the low-frequency band and high-frequency band, respectively. So, taking loudness and sharpness as the core indicators under accelerating conditions is recommended. Then, the sound quality of the ETCC under unstable conditions at variable speeds is investigated. It is found that the sound quality of the ETCC deteriorates under surge conditions. When entering into the deep surge, the loudness and sharpness both fluctuate drastically, which is caused by the periodic impulse noise. The fluctuation strength rises sharply and keeps a high level in the process of deep surge because of the obvious low-frequency characteristics of the surge. The above three parameters have a clear correlation with the rotational speed, while roughness is the only parameter that is not significantly connected with the rotational speed, which can be explained by the existence of a high-frequency fluctuation under the entire surge process. Finally, the causes of sound quality deterioration are explored using secondary signal processing. The results show that the outlet pressure pulsations of the ETCC under surge conditions lead to the increases and fluctuations of these acoustic parameters, and compressor surge is the main cause of deterioration in sound quality for FCVs. Based on the results, the objective sound quality evaluation of the ETCC is accomplished, and four psychoacoustic parameters can be modified to improve the sound quality. This study provides a theoretical basis for proposing methods to accurately evaluate and improve the sound quality of FCVs.

Research on sound quality of roller chain transmission system based on multi-source transfer learning

To establish the sound quality evaluation model of roller chain transmission system, we collect the running noise under different working conditions. After the noise samples are preprocessed, a group of experienced testers are organized to evaluate them subjectively. Mel frequency cepstral coefficient (MFCC) of each noise sample is calculated, and the MFCC feature map is used as an objective evaluation. Combining with the subjective and objective evaluation results of the roller chain system noise, we can get the original dataset of its sound quality research. However, the number of high-quality noise samples is relatively small. Based on the sound quality research of various chain transmission systems, a novel method called multi-source transfer learning convolutional neural network (MSTL-CNN) is proposed. By transferring knowledge from multiple source tasks to target task, the difficulty of small sample sound quality prediction is solved. Compared with the problem that single source task transfer learning has too much error on some samples, MSTL-CNN can give full play to the advantages of all transfer learning models. The results also show that the MSTL-CNN proposed in this paper is significantly better than the traditional sound quality evaluation methods.

Associations between illness perceptions, word recognition, and perceived sound quality in cochlear implant users.

Illness perceptions refer to thoughts and ideas an individual has about an illness. The aim was to understand how cochlear implant (CI) users' illness perceptions, in addition to their monosyllabic word recognition abilities, are associated with their self-perceived sound quality. Data were collected during routine CI check-up appointments. Participants completed the Brief Illness Perception Questionnaire (assessing their illness perceptions) and the Hearing Implant Sound Quality Index (assessing their subjective sound quality). Additionally, monosyllabic word recognition abilities were measured with the Freiburg Monosyllable Word Test. Hierarchical regression analysis were utilised to model users' sound quality ratings. Participants' age was entered first as a control variable. In the next step, monosyllabic word recognition was entered. Finally, participants' illness perceptions were entered. Fifty-five participants with unilateral CI provision. Monosyllabic word recognition was significant in the second step. When illness perceptions and monosyllabic word recognition were both included in the third step, illness perceptions, but not monosyllabic word recognition, were significant. The model explained 22% of the variance of subjective sound quality. Monosyllabic word recognition abilities and illness perceptions of CI users are important for their self-reported sound quality, but illness perceptions appear to be potentially more relevant.

Analysis of Artistic Instruction and Emotional Expression Pathways in College Piano Performance in the Internet Era

Abstract As computer science advances, it intersects intriguingly with the realm of music acoustics, particularly in enhancing piano performance through technological means. This paper delves into an innovative approach to piano learning and creation, focusing on emotional expression’s nuances. We have devised a system capable of precise musical tone recognition and sound quality evaluation by adopting Mel Frequency Cepstral Coefficients (MFCC) for the nuanced extraction of piano sounds and integrating dynamic fuzzy neural networks. Our findings show an impressive accuracy rate, with musical tone misidentification below 2.58% and sound quality assessment errors within a 5% margin. This work not only sets a new benchmark in piano performance analysis but also paves the way for revolutionary teaching methods in music education, with profound implications for artistic instruction and emotional expression.

Mechanisms for the application of harmonic theory in enhancing choral sound quality

Abstract The development of traditional music is facing the change of digitalization and multimedia, and the application of research and acoustic theory in enhancing the effect of choral music with the help of computer technology is an important trend in the development of musicology at present. The article first proposes a method for choral training and then combines the LCQA algorithm and the FBE algorithm based on Mel filtering energy to propose a method for evaluating the quality of choral sound. The article finally takes a chorus as an empirical research object. It conducts specific experimental tests using the method proposed in this article, and the experimental results obtained are, in the analysis of choral sound effect, the total hoarseness and roughness of choral sound have decreased by about 0.8±0.31. In the evaluation of choral sound quality, the average score of the overall sound quality has risen by 23.25 points.

Taut-MUSHRA: A MUSHRA-based method without hidden reference and anchors for relative sound quality evaluation

Taut-MUSHRA: A MUSHRA-based method without hidden reference and anchors for relative sound quality evaluation

A rapid evaluation method for the quality consistency assessment and spectrum-effect relationship study of Xiaohuoluo Pills developed based on the spectral and chromatography combined technology

To ensure the effectiveness of traditional Chinese medicine (TCM) and drive the advancement of high-quality Chinese medicine, the establishment of a scientifically sound quality evaluation system is particularly important. The...

Shin-Sound Quality Evaluation: Image recognition AI for sound quality evaluation

Sound quality evaluation are historically based on psychoacoustical parameters, namely loudness, sharpness and so on. From awake of massive progress of Artificial Intelligence, analysis of sound should be affected by this novel technology. Therefore, in this paper the author introduces an approach with image recognition AI into sound quality evaluation. It results in showing advantages over complex and non-linear sound evaluation. Thus, the idea of "eyes to ears" could lead new application to this type of solutions.

Quantification of sound quality for lawnmower radiated noise considering listening conditions

In this study, we investigated an influence of listening conditions on the sound quality evaluation of lawnmower radiated noise and attempted to quantify the quality. Subjective evaluation tests were conducted using various lawnmowers equipped with engine or electric motor. In the active listening condition test, the participants operated the lawnmower by themselves and evaluated "discomfort" and "powerful" of the sound. They again evaluated the sound quality of the identical reproduced sound in a test room in the passive test. As the results, discomfort of the sound was significantly increased in the passive condition although powerful was almost same. Subsequently, we attempted to quantify these sound quality by using psychoacoustic metrics in each condition. The result indicated that both sound quality could be expressed very well using loudness and sharpness. Loud and high pitch sound was evaluated as more discomfort and loud and low pitch sound was more powerful. In addition, discomfort of the sound was clarified to be evaluated about 30% more in the passive condition than that in the active condition.