Applied Acoustics最新文献_第7页

Underwater acoustic classification using wavelet scattering transform and convolutional neural network with limited dataset

IF 3.4 2区物理与天体物理 Q1 ACOUSTICS

Applied Acoustics

Pub Date : 2025-01-31 DOI: 10.1016/j.apacoust.2025.110564

Yongxiang Liu , Biqi Zhang , Fantong Kong , Biao Wang , Chengming Luo , Lin Ma

Underwater acoustic signal classification plays a pivotal role in maritime applications, requiring accurately identifying various acoustic sources in complex underwater environments. While deep learning has substantially enhanced performance in this domain, its success is often contingent on hand-crafted input features and intricate network architectures. The paper presents a novel method for classifying underwater acoustic signals by integrating the Wavelet Scattering Transform (WST) with Attention-augmented Convolutional Neural Networks (CNNs). The WST, based on wavelet analysis, effectively extracts multiscale features while retaining crucial time-frequency information, offering translation invariance and reducing the dependency on large training datasets. Furthermore, incorporating ResNet-18 with an attention mechanism improves extracted features by capturing richer semantic information, even from limited training data. The method was evaluated on the ShipsEar dataset, utilizing only 8.5% of samples for training, 1.5% for validation, and the remaining 90% for testing. Our approach achieved a classification accuracy of 0.93, surpassing the traditional Mel spectrogram with ResNet-18 by 9.8%. These results underscore the effectiveness of the proposed method in handling challenging underwater acoustic environments with limited training data.

{"title":"Underwater acoustic classification using wavelet scattering transform and convolutional neural network with limited dataset","authors":"Yongxiang Liu , Biqi Zhang , Fantong Kong , Biao Wang , Chengming Luo , Lin Ma","doi":"10.1016/j.apacoust.2025.110564","DOIUrl":"10.1016/j.apacoust.2025.110564","url":null,"abstract":"<div><div>Underwater acoustic signal classification plays a pivotal role in maritime applications, requiring accurately identifying various acoustic sources in complex underwater environments. While deep learning has substantially enhanced performance in this domain, its success is often contingent on hand-crafted input features and intricate network architectures. The paper presents a novel method for classifying underwater acoustic signals by integrating the Wavelet Scattering Transform (WST) with Attention-augmented Convolutional Neural Networks (CNNs). The WST, based on wavelet analysis, effectively extracts multiscale features while retaining crucial time-frequency information, offering translation invariance and reducing the dependency on large training datasets. Furthermore, incorporating ResNet-18 with an attention mechanism improves extracted features by capturing richer semantic information, even from limited training data. The method was evaluated on the ShipsEar dataset, utilizing only 8.5% of samples for training, 1.5% for validation, and the remaining 90% for testing. Our approach achieved a classification accuracy of 0.93, surpassing the traditional Mel spectrogram with ResNet-18 by 9.8%. These results underscore the effectiveness of the proposed method in handling challenging underwater acoustic environments with limited training data.</div></div>","PeriodicalId":55506,"journal":{"name":"Applied Acoustics","volume":"232 ","pages":"Article 110564"},"PeriodicalIF":3.4,"publicationDate":"2025-01-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143261696","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Multi-gradient acoustic black hole metamaterial for near-perfect sound Attenuation: Theory, simulation and experiments

IF 3.4 2区物理与天体物理 Q1 ACOUSTICS

Applied Acoustics

Pub Date : 2025-01-25 DOI: 10.1016/j.apacoust.2025.110546

Xinhao Zhang , Mingjing Geng , Caiyou Zhao , Yajun Cao , Ping Wang

A multi-gradient acoustic black hole metamaterial module (MABHM) based on a genetic algorithm-neural network (NN-GA) is optimally designed for near-perfect sound attenuation of sub-wavelength metamaterial structures. Simulation results show that the absorption coefficient of the MABHM is up to more than 0.9 in the range of 300 Hz-20 kHz, and the MABHM has a good absorption effect for sound waves with different incident angles. The complex acoustic impedance of the MABHM has a phase loop approximating near the (1,0) coordinates. It is proved that the impedance matching effect is the key to realizing near-perfect sound absorption. The sound transmission loss curves of MABHM with different perforation ratios are the same, and the sound insulation at frequencies above 200 Hz reaches more than 35.0 dB. The experimental results of sound absorption coefficient and sound transmission loss are consistent with the simulation results. MABHM has the advantages of low-frequency broadband sound absorption and insulation, which can be used to realize nearly perfect noise reduction.

{"title":"Multi-gradient acoustic black hole metamaterial for near-perfect sound Attenuation: Theory, simulation and experiments","authors":"Xinhao Zhang , Mingjing Geng , Caiyou Zhao , Yajun Cao , Ping Wang","doi":"10.1016/j.apacoust.2025.110546","DOIUrl":"10.1016/j.apacoust.2025.110546","url":null,"abstract":"<div><div>A multi-gradient acoustic black hole metamaterial module (MABHM) based on a genetic algorithm-neural network (NN-GA) is optimally designed for near-perfect sound attenuation of sub-wavelength metamaterial structures. Simulation results show that the absorption coefficient of the MABHM is up to more than 0.9 in the range of 300 Hz-20 kHz, and the MABHM has a good absorption effect for sound waves with different incident angles. The complex acoustic impedance of the MABHM has a phase loop approximating near the (1,0) coordinates. It is proved that the impedance matching effect is the key to realizing near-perfect sound absorption. The sound transmission loss curves of MABHM with different perforation ratios are the same, and the sound insulation at frequencies above 200 Hz reaches more than 35.0 dB. The experimental results of sound absorption coefficient and sound transmission loss are consistent with the simulation results. MABHM has the advantages of low-frequency broadband sound absorption and insulation, which can be used to realize nearly perfect noise reduction.</div></div>","PeriodicalId":55506,"journal":{"name":"Applied Acoustics","volume":"231 ","pages":"Article 110546"},"PeriodicalIF":3.4,"publicationDate":"2025-01-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143095399","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Research on the effect of auditory and visual environmental factors on patients’ waiting experience in healthcare

IF 3.4 2区物理与天体物理 Q1 ACOUSTICS

Applied Acoustics

Pub Date : 2025-01-25 DOI: 10.1016/j.apacoust.2025.110540

Tianfu Zhou, Qin Wei, Kun Liu

Background

Implementing visual and auditory stimuli is recognized as an effective strategy to improve patients’ waiting experiences. However, the impact of window size and sound category on psychological responses has received scant attention. Besides, there is a lack of audio-visual interaction research in healthcare settings.

Objectives

This experimental study investigated the main and interactive effects of window size and sound category on occupants’ emotional state, environmental evaluation, and perceived waiting time.

Methods

In this study, a between-subjects laboratory experiment was conducted utilizing virtual reality. A total of 387 subjects were randomly assigned to one of 16 experiment conditions, which were combinations of four different window size scenarios and 4 distinct sound categories, to test the effects of window size and sound type as well as their interactions.

Results

The statistical findings indicated that both the visual factor (the window size) and the auditory factor (the sound category) exert statistically significant influences on psychological indicators. Moreover, a significant interaction effect between the visual and auditory factors was observed on subjects’ emotional states. Notably, mechanical sound was found to counteract the restorative effects of window, whereas increasing window size in anthropogenic sound can lead to enhanced emotional benefits.

Conclusions

The study provides insight into the role of window size and sound category in affecting occupants’ waiting experience, emphasizing the interactive effects between visual and auditory elements. It is anticipated that the research will contribute to the multisensory literature and provide evidence-based guidance for the design of restorative waiting environments.

{"title":"Research on the effect of auditory and visual environmental factors on patients’ waiting experience in healthcare","authors":"Tianfu Zhou, Qin Wei, Kun Liu","doi":"10.1016/j.apacoust.2025.110540","DOIUrl":"10.1016/j.apacoust.2025.110540","url":null,"abstract":"<div><h3>Background</h3><div>Implementing visual and auditory stimuli is recognized as an effective strategy to improve patients’ waiting experiences. However, the impact of window size and sound category on psychological responses has received scant attention. Besides, there is a lack of audio-visual interaction research in healthcare settings.</div></div><div><h3>Objectives</h3><div>This experimental study investigated the main and interactive effects of window size and sound category on occupants’ emotional state, environmental evaluation, and perceived waiting time.</div></div><div><h3>Methods</h3><div>In this study, a between-subjects laboratory experiment was conducted utilizing virtual reality. A total of 387 subjects were randomly assigned to one of 16 experiment conditions, which were combinations of four different window size scenarios and 4 distinct sound categories, to test the effects of window size and sound type as well as their interactions.</div></div><div><h3>Results</h3><div>The statistical findings indicated that both the visual factor (the window size) and the auditory factor (the sound category) exert statistically significant influences on psychological indicators. Moreover, a significant interaction effect between the visual and auditory factors was observed on subjects’ emotional states. Notably, mechanical sound was found to counteract the restorative effects of window, whereas increasing window size in anthropogenic sound can lead to enhanced emotional benefits.</div></div><div><h3>Conclusions</h3><div>The study provides insight into the role of window size and sound category in affecting occupants’ waiting experience, emphasizing the interactive effects between visual and auditory elements. It is anticipated that the research will contribute to the multisensory literature and provide evidence-based guidance for the design of restorative waiting environments.</div></div>","PeriodicalId":55506,"journal":{"name":"Applied Acoustics","volume":"231 ","pages":"Article 110540"},"PeriodicalIF":3.4,"publicationDate":"2025-01-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143095403","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Analysing the bayan drum; its tuning, and its bols from a musical standpoint

IF 3.4 2区物理与天体物理 Q1 ACOUSTICS

Applied Acoustics

Pub Date : 2025-01-24 DOI: 10.1016/j.apacoust.2025.110536

Sreerag Ashok, Nachiketa Tiwari

The Indian tabla is a pair of drums; the bayan played with the left hand, and the dayan played with the right hand. The bayan is often played in ways such that its pitch changes during a performance. Past research on the bayan is limited and has primarily focused on the eigenmodes and eigenvalues of the drum’s membrane, and the harmonicity of the drum. What has not been explored is the nature of its bols (musical sounds) and how they relate to the drum’s design and its playing technique. There is also a need to understand the acoustical basis underlying its tuning process, which can often be a time-consuming exercise. We have attempted to fill such gaps by conducting a detailed experimental study of the bayan. Towards such a goal, we have captured the displacement of the bayan membrane and the resulting audio signal of musical bols emanating from the bayan. For capturing membrane displacements, we employed two stereoscopically arranged high-speed cameras and the Digital Image Correlation (DIC) techniques along with a high-speed data acquisition system. Our investigations revealed how the presence of asymmetry in membrane tension can result in frequency splitting around the second harmonic frequency, the variation of corresponding peak amplitudes due to the rotation of striking position, and the phenomenon of degenerate modes around the third harmonic frequency. We also found a correlation between the playing technique of the /ɡʰe/ bol, its decay time and the extent of variation in the bayan’s lowest frequency. Our work also revealed linkages between several other spectral features of the /ɡaː/ and /ɡʰe/ bols and the tabla’s structure and its playing technique. All such understandings can help us design a better bayan and a detuning detector, which can speed up the bayan tuning process.

{"title":"Analysing the bayan drum; its tuning, and its bols from a musical standpoint","authors":"Sreerag Ashok, Nachiketa Tiwari","doi":"10.1016/j.apacoust.2025.110536","DOIUrl":"10.1016/j.apacoust.2025.110536","url":null,"abstract":"<div><div>The Indian tabla is a pair of drums; the <em>bayan</em> played with the left hand, and the <em>dayan</em> played with the right hand. The <em>bayan</em> is often played in ways such that its pitch changes during a performance. Past research on the <em>bayan</em> is limited and has primarily focused on the eigenmodes and eigenvalues of the drum’s membrane, and the harmonicity of the drum. What has not been explored is the nature of its <em>bols</em> (musical sounds) and how they relate to the drum’s design and its playing technique. There is also a need to understand the acoustical basis underlying its tuning process, which can often be a time-consuming exercise. We have attempted to fill such gaps by conducting a detailed experimental study of the <em>bayan</em>. Towards such a goal, we have captured the displacement of the <em>bayan</em> membrane and the resulting audio signal of musical <em>bols</em> emanating from the <em>bayan</em>. For capturing membrane displacements, we employed two stereoscopically arranged high-speed cameras and the Digital Image Correlation (DIC) techniques along with a high-speed data acquisition system. Our investigations revealed how the presence of asymmetry in membrane tension can result in frequency splitting around the second harmonic frequency, the variation of corresponding peak amplitudes due to the rotation of striking position, and the phenomenon of degenerate modes around the third harmonic frequency. We also found a correlation between the playing technique of the /ɡʰe/ <em>bol</em>, its decay time and the extent of variation in the <em>bayan</em>’s lowest frequency. Our work also revealed linkages between several other spectral features of the /ɡaː/ and /ɡʰe/ <em>bols</em> and the tabla’s structure and its playing technique. All such understandings can help us design a better <em>bayan</em> and a detuning detector, which can speed up the <em>bayan</em> tuning process.</div></div>","PeriodicalId":55506,"journal":{"name":"Applied Acoustics","volume":"231 ","pages":"Article 110536"},"PeriodicalIF":3.4,"publicationDate":"2025-01-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143095404","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A Synchronized Filter-s Least Mean Square (SFsLMS) algorithm for multi-channel ANC in aviation noise suppression

IF 3.4 2区物理与天体物理 Q1 ACOUSTICS

Applied Acoustics

Pub Date : 2025-01-23 DOI: 10.1016/j.apacoust.2025.110552

Mossalam Khamis, Sheng Zhang, Saeed Ibrahim

This paper presents a synchronized Filtered-s Least Mean Squares (SFsLMS) algorithm for multichannel Active Noise Control (ANC) systems aimed at mitigating aviation noise. The SFsLMS algorithm addresses signal delays inherent in aircraft environments, which degrade the performance of traditional ANC algorithms. Incorporating delay estimation into the adaptive filtering process ensures accurate alignment of input and reference signals, leading to improved convergence speed and stability. The results demonstrate that the SFsLMS algorithm significantly enhances noise cancellation performance in dynamic aviation noise conditions, offering a scalable and robust solution for real-time noise reduction in enclosed areas near airports. This advancement contributes to increased comfort and reduced noise pollution, highlighting the algorithm's potential for widespread application in aviation noise control systems. The evaluation is conducted using a

(2 \times 4 \times 4)

(ANC) system, with performance measured in terms of Averaged Noise Reduction (ANR). The results reveal a marked improvement in convergence speed and stability, as demonstrated by the rapid decrease and sustained low levels of ANR across all microphones.

{"title":"A Synchronized Filter-s Least Mean Square (SFsLMS) algorithm for multi-channel ANC in aviation noise suppression","authors":"Mossalam Khamis, Sheng Zhang, Saeed Ibrahim","doi":"10.1016/j.apacoust.2025.110552","DOIUrl":"10.1016/j.apacoust.2025.110552","url":null,"abstract":"<div><div>This paper presents a synchronized Filtered-s Least Mean Squares (SFsLMS) algorithm for multichannel Active Noise Control (ANC) systems aimed at mitigating aviation noise. The SFsLMS algorithm addresses signal delays inherent in aircraft environments, which degrade the performance of traditional ANC algorithms. Incorporating delay estimation into the adaptive filtering process ensures accurate alignment of input and reference signals, leading to improved convergence speed and stability. The results demonstrate that the SFsLMS algorithm significantly enhances noise cancellation performance in dynamic aviation noise conditions, offering a scalable and robust solution for real-time noise reduction in enclosed areas near airports. This advancement contributes to increased comfort and reduced noise pollution, highlighting the algorithm's potential for widespread application in aviation noise control systems. The evaluation is conducted using a <span><math><mo>(</mo><mn>2</mn><mo>×</mo><mn>4</mn><mo>×</mo><mn>4</mn><mo>)</mo></math></span> (ANC) system, with performance measured in terms of Averaged Noise Reduction (ANR). The results reveal a marked improvement in convergence speed and stability, as demonstrated by the rapid decrease and sustained low levels of ANR across all microphones.</div></div>","PeriodicalId":55506,"journal":{"name":"Applied Acoustics","volume":"231 ","pages":"Article 110552"},"PeriodicalIF":3.4,"publicationDate":"2025-01-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143095405","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

An acoustic survey of Korean school classrooms: The necessity for Korean acoustic standards and design guideleines

IF 3.4 2区物理与天体物理 Q1 ACOUSTICS

Applied Acoustics

Pub Date : 2025-01-22 DOI: 10.1016/j.apacoust.2025.110548

Young-Ji Choi

This paper presents the results of a recent noise and acoustic survey conducted in school classrooms in Korea. The objective of the survey was to provide information on the acoustic characteristics of classrooms in Korea and to emphasize the necessity of introducing acoustic standards for classrooms. The survey includes unoccupied data on the acoustic conditions and sound insulation performance in addition to occupied speech and noise levels in 16 classrooms in four schools with a range of student ages. None of the classrooms complied with the background noise level, reverberation time, and airborne sound insulation requirements aa defined in the ANSI and BB93 standards. All eight classrooms were found to meet the impact sound insulation requirements. A Gaussian mixture modelling analysis of the data from all 27 classes in the four schools revealed that the mean speech level was 65.1 dBA (s.d. = 5.2), the mean noise level was 50 dBA (s.d. = 6.1), and the difference between the speech and noise levels was 15.1 dBA (s.d. = 3.3). It was observed that the mean speech levels were higher in elementary school classrooms than in junior high, high, and special school classrooms, with a parallel trend evident in noise levels. The findings indicate a general trend of decreasing noise levels with increasing age of the students and decreasing numbers of students. A significant correlation was observed between speech and student activity noise levels, with a 0.71 dBA increase in speech levels for every 1 dBA in noise levels. Furthermore, the noise levels in the school classrooms were found to be closely related to the type of classroom activity. The predicted T₃₀ values for occupied conditions showed a statistically significant correlation with the noise levels and SNR values. A significant correlation was observed between the mean noise levels in classrooms and noise levels in corridors, indicating that the management of disruptive noise from adjacent spaces is essential to ensure optimal acoustics for speech in classrooms.

{"title":"An acoustic survey of Korean school classrooms: The necessity for Korean acoustic standards and design guideleines","authors":"Young-Ji Choi","doi":"10.1016/j.apacoust.2025.110548","DOIUrl":"10.1016/j.apacoust.2025.110548","url":null,"abstract":"<div><div>This paper presents the results of a recent noise and acoustic survey conducted in school classrooms in Korea. The objective of the survey was to provide information on the acoustic characteristics of classrooms in Korea and to emphasize the necessity of introducing acoustic standards for classrooms. The survey includes unoccupied data on the acoustic conditions and sound insulation performance in addition to occupied speech and noise levels in 16 classrooms in four schools with a range of student ages. None of the classrooms complied with the background noise level, reverberation time, and airborne sound insulation requirements aa defined in the ANSI and BB93 standards. All eight classrooms were found to meet the impact sound insulation requirements. A Gaussian mixture modelling analysis of the data from all 27 classes in the four schools revealed that the mean speech level was 65.1 dBA (s.d. = 5.2), the mean noise level was 50 dBA (s.d. = 6.1), and the difference between the speech and noise levels was 15.1 dBA (s.d. = 3.3). It was observed that the mean speech levels were higher in elementary school classrooms than in junior high, high, and special school classrooms, with a parallel trend evident in noise levels. The findings indicate a general trend of decreasing noise levels with increasing age of the students and decreasing numbers of students. A significant correlation was observed between speech and student activity noise levels, with a 0.71 dBA increase in speech levels for every 1 dBA in noise levels. Furthermore, the noise levels in the school classrooms were found to be closely related to the type of classroom activity. The predicted T<sub>30</sub> values for occupied conditions showed a statistically significant correlation with the noise levels and SNR values. A significant correlation was observed between the mean noise levels in classrooms and noise levels in corridors, indicating that the management of disruptive noise from adjacent spaces is essential to ensure optimal acoustics for speech in classrooms.</div></div>","PeriodicalId":55506,"journal":{"name":"Applied Acoustics","volume":"231 ","pages":"Article 110548"},"PeriodicalIF":3.4,"publicationDate":"2025-01-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143095406","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

High-order similarity learning based domain adaptation for speech emotion recognition

IF 3.4 2区物理与天体物理 Q1 ACOUSTICS

Applied Acoustics

Pub Date : 2025-01-22 DOI: 10.1016/j.apacoust.2025.110555

Hao Wang , Yixuan Ji , Peng Song , Zhaowei Liu

Speech emotion recognition (SER) has received significant attention due to the advancement of artificial intelligence technology. Conventional SER methods usually assume that both the training and test data are derived from the same dataset, without fully considering the differences between different datasets, which would lead to reduced recognition performance. To address this problem, this paper proposes a novel domain adaptation approach called high-order similarity learning based domain adaptation (HSDA) for SER. Specifically, we first project the original data into a low-dimensional embedding subspace, which can effectively eliminate the inter-domain differences. Then, we learn the high-order similarity graph to exploit the intrinsic structural information of cross-domain data. At the same time, we utilize the regression term to enhance the discriminative power of the model, which can fully use the labeling information of the source domain to make the learned transformation matrix more discriminative. The experimental results on four popular datasets show that our method can achieve excellent performance compared to several state-of-the-art methods.

{"title":"High-order similarity learning based domain adaptation for speech emotion recognition","authors":"Hao Wang , Yixuan Ji , Peng Song , Zhaowei Liu","doi":"10.1016/j.apacoust.2025.110555","DOIUrl":"10.1016/j.apacoust.2025.110555","url":null,"abstract":"<div><div>Speech emotion recognition (SER) has received significant attention due to the advancement of artificial intelligence technology. Conventional SER methods usually assume that both the training and test data are derived from the same dataset, without fully considering the differences between different datasets, which would lead to reduced recognition performance. To address this problem, this paper proposes a novel domain adaptation approach called high-order similarity learning based domain adaptation (HSDA) for SER. Specifically, we first project the original data into a low-dimensional embedding subspace, which can effectively eliminate the inter-domain differences. Then, we learn the high-order similarity graph to exploit the intrinsic structural information of cross-domain data. At the same time, we utilize the regression term to enhance the discriminative power of the model, which can fully use the labeling information of the source domain to make the learned transformation matrix more discriminative. The experimental results on four popular datasets show that our method can achieve excellent performance compared to several state-of-the-art methods.</div></div>","PeriodicalId":55506,"journal":{"name":"Applied Acoustics","volume":"231 ","pages":"Article 110555"},"PeriodicalIF":3.4,"publicationDate":"2025-01-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143095407","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Research on the multi-signal DOA estimation based on ResNet with the attention module combined with beamforming (RAB-DOA)

IF 3.4 2区物理与天体物理 Q1 ACOUSTICS

Applied Acoustics

Pub Date : 2025-01-21 DOI: 10.1016/j.apacoust.2025.110541

Long Wu , Yue Fu , Xu Yang , Lu Xu , Shuyu Chen , Yong Zhang , Jianlong Zhang

Direction of Arrival (DOA) estimation based on deep neural networks has been extensively studied recently, but multi-signal DOA estimation has not been sufficiently investigated. The strong mutual interference between signals emitted by multiple sources in different directions in multiple DOA leads to the reduction of detection accuracy, which limits the application in multi-object scenarios. In multi-signal DOA estimation, a residual network (ResNet) incorporating efficient channel attention module could significantly enhance the signal separation and localisation capabilities of the system. Therefore, this paper presents a multi-signal DOA estimation system based on ResNet with the attention module and beamforming (RAB-DOA). The system receives spatial signals through an array of detectors and uses a linear constrained minimum variance (LCMV) beamforming algorithm to optimize signal directivity and suppress interference. Phase adjustment is then performed during the scanning process to enhance the signal in the scanning direction and suppress interfering signals in other directions. Finally, the signals are binary classified using ResNet with an efficient channel attention module to obtain multi-signal DOA estimation results. Experiment results show that the detection accuracy and precision of the proposed algorithm are excellent, especially at low SNRs in spite of multiple interfering signals.

{"title":"Research on the multi-signal DOA estimation based on ResNet with the attention module combined with beamforming (RAB-DOA)","authors":"Long Wu , Yue Fu , Xu Yang , Lu Xu , Shuyu Chen , Yong Zhang , Jianlong Zhang","doi":"10.1016/j.apacoust.2025.110541","DOIUrl":"10.1016/j.apacoust.2025.110541","url":null,"abstract":"<div><div>Direction of Arrival (DOA) estimation based on deep neural networks has been extensively studied recently, but multi-signal DOA estimation has not been sufficiently investigated. The strong mutual interference between signals emitted by multiple sources in different directions in multiple DOA leads to the reduction of detection accuracy, which limits the application in multi-object scenarios. In multi-signal DOA estimation, a residual network (ResNet) incorporating efficient channel attention module could significantly enhance the signal separation and localisation capabilities of the system. Therefore, this paper presents a multi-signal DOA estimation system based on ResNet with the attention module and beamforming (RAB-DOA). The system receives spatial signals through an array of detectors and uses a linear constrained minimum variance (LCMV) beamforming algorithm to optimize signal directivity and suppress interference. Phase adjustment is then performed during the scanning process to enhance the signal in the scanning direction and suppress interfering signals in other directions. Finally, the signals are binary classified using ResNet with an efficient channel attention module to obtain multi-signal DOA estimation results. Experiment results show that the detection accuracy and precision of the proposed algorithm are excellent, especially at low SNRs in spite of multiple interfering signals.</div></div>","PeriodicalId":55506,"journal":{"name":"Applied Acoustics","volume":"231 ","pages":"Article 110541"},"PeriodicalIF":3.4,"publicationDate":"2025-01-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143095921","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Prior knowledge-guided multi-scale acoustic metamaterial sensing for gearbox weak fault signal detection

IF 3.4 2区物理与天体物理 Q1 ACOUSTICS

Applied Acoustics

Pub Date : 2025-01-18 DOI: 10.1016/j.apacoust.2025.110532

Yaqin Wang , Jia Liu , Huafei Pan , Zhao Huang , Jiaowei Xiao , Xiaoxi Ding

The early fault detection presents a significant challenge due to the intricate structure of the gearbox, substantial noise interference, and multi-component coupling modulation. Traditional post-processing algorithms are relatively complex and inefficient. Motivated by the properties of acoustic metamaterial in feature enhancement and amplitude-frequency modulation mechanism of signal processing, this study proposes multi-scale acoustic metamaterials (MSAM) for gearbox weak fault signal detection with multi-scale feature information synthesized. Specially, benefiting from the merits of acoustic rainbow capture in amplitude gain and noise suppression, this front-end enhanced sensing approach exploits the properties of acoustic compression and feature separation of different frequency components of sound waves. Guided by prior knowledge of gearbox modulation mechanisms, the acoustic metamaterial structure is firstly optimized and miniaturized, followed by experimental testing of the center frequency and bandwidth of each air gap. Notably, the single air gap of this designed MSAM is verified that an amplitude gain exceeding 10 times for target components at a single scale can be achieved according to the results of fault simulation signal testing. Thereupon, focusing on issue of multi-scale coupling modulation, two cases has been also provided to illustrate the ability of multi-scale feature extraction with three adjacent air gaps and two non-adjacent gaps from MSAM. These indicate that the proposed front-end enhanced sensing structure can provide a more comprehensive and distinct representation than that of fault characteristics obtained from free-field collected signals even under strong noise and complex multi-scale coupling interferences. It can be foreseen that the proposed mechanical signal sensing driven with acoustic metamaterial brings great potential in weak signal detection, and it also shows the expectation of achieving variable scale adaptive control and material intelligent sensing.

{"title":"Prior knowledge-guided multi-scale acoustic metamaterial sensing for gearbox weak fault signal detection","authors":"Yaqin Wang , Jia Liu , Huafei Pan , Zhao Huang , Jiaowei Xiao , Xiaoxi Ding","doi":"10.1016/j.apacoust.2025.110532","DOIUrl":"10.1016/j.apacoust.2025.110532","url":null,"abstract":"<div><div>The early fault detection presents a significant challenge due to the intricate structure of the gearbox, substantial noise interference, and multi-component coupling modulation. Traditional post-processing algorithms are relatively complex and inefficient. Motivated by the properties of acoustic metamaterial in feature enhancement and amplitude-frequency modulation mechanism of signal processing, this study proposes multi-scale acoustic metamaterials (MSAM) for gearbox weak fault signal detection with multi-scale feature information synthesized. Specially, benefiting from the merits of acoustic rainbow capture in amplitude gain and noise suppression, this front-end enhanced sensing approach exploits the properties of acoustic compression and feature separation of different frequency components of sound waves. Guided by prior knowledge of gearbox modulation mechanisms, the acoustic metamaterial structure is firstly optimized and miniaturized, followed by experimental testing of the center frequency and bandwidth of each air gap. Notably, the single air gap of this designed MSAM is verified that an amplitude gain exceeding 10 times for target components at a single scale can be achieved according to the results of fault simulation signal testing. Thereupon, focusing on issue of multi-scale coupling modulation, two cases has been also provided to illustrate the ability of multi-scale feature extraction with three adjacent air gaps and two non-adjacent gaps from MSAM. These indicate that the proposed front-end enhanced sensing structure can provide a more comprehensive and distinct representation than that of fault characteristics obtained from free-field collected signals even under strong noise and complex multi-scale coupling interferences. It can be foreseen that the proposed mechanical signal sensing driven with acoustic metamaterial brings great potential in weak signal detection, and it also shows the expectation of achieving variable scale adaptive control and material intelligent sensing.</div></div>","PeriodicalId":55506,"journal":{"name":"Applied Acoustics","volume":"231 ","pages":"Article 110532"},"PeriodicalIF":3.4,"publicationDate":"2025-01-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143095923","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Design of subwavelength wide bandwidth sound absorbers by inverse convolutional neural networks

IF 3.4 2区物理与天体物理 Q1 ACOUSTICS

Applied Acoustics

Pub Date : 2025-01-18 DOI: 10.1016/j.apacoust.2025.110543

Peter Hawes, Marco Boccaccio, Michele Meo

Microperforated panel sound absorber metamaterials are crucial for noise reduction in various applications. This study leverages a convolutional neural network (CNN) machine learning model to optimise these metamaterials for maximum absorption strength and bandwidth range. The model allows for inverse optimisation of sound absorption performance. A desired absorption response can be supplied as input, and the network returns the necessary geometry parameters to achieve the target characteristic.

Metamaterials were optimised to provide over 90 % absorption at target frequencies between 0–1000 Hz. Theoretical predictions were validated experimentally via impedance tube testing. The model achieved no less than 70 % absorption over a 923 Hz range (548–1471 Hz) with a material thickness of 41 mm, and 70 % absorption over 1000 Hz (470–1470 Hz) with a thickness of 57 mm. A case study for an automotive/energy application targeted 50 % absorption between 500–1000 Hz at a thickness of less than 25 mm. Experimental results showed 50 % absorption between 506–1032 Hz at 23 mm thickness. These findings demonstrate the potential of CNN models in optimising sound absorber metamaterials, offering significant improvements in noise reduction with minimal material thickness. The proposed methodology offers significant potential for lightweight applications in various noise-reduction scenarios, including automotive, aerospace, energy, and architectural acoustics.

{"title":"Design of subwavelength wide bandwidth sound absorbers by inverse convolutional neural networks","authors":"Peter Hawes, Marco Boccaccio, Michele Meo","doi":"10.1016/j.apacoust.2025.110543","DOIUrl":"10.1016/j.apacoust.2025.110543","url":null,"abstract":"<div><div>Microperforated panel sound absorber metamaterials are crucial for noise reduction in various applications. This study leverages a convolutional neural network (CNN) machine learning model to optimise these metamaterials for maximum absorption strength and bandwidth range. The model allows for inverse optimisation of sound absorption performance. A desired absorption response can be supplied as input, and the network returns the necessary geometry parameters to achieve the target characteristic.</div><div>Metamaterials were optimised to provide over 90 % absorption at target frequencies between 0–1000 Hz. Theoretical predictions were validated experimentally via impedance tube testing. The model achieved no less than 70 % absorption over a 923 Hz range (548–1471 Hz) with a material thickness of 41 mm, and 70 % absorption over 1000 Hz (470–1470 Hz) with a thickness of 57 mm. A case study for an automotive/energy application targeted 50 % absorption between 500–1000 Hz at a thickness of less than 25 mm. Experimental results showed 50 % absorption between 506–1032 Hz at 23 mm thickness. These findings demonstrate the potential of CNN models in optimising sound absorber metamaterials, offering significant improvements in noise reduction with minimal material thickness. The proposed methodology offers significant potential for lightweight applications in various noise-reduction scenarios, including automotive, aerospace, energy, and architectural acoustics.</div></div>","PeriodicalId":55506,"journal":{"name":"Applied Acoustics","volume":"231 ","pages":"Article 110543"},"PeriodicalIF":3.4,"publicationDate":"2025-01-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143095925","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0