Applied Acoustics最新文献

{"title":"High-order similarity learning based domain adaptation for speech emotion recognition","authors":"Hao Wang ,&nbsp;Yixuan Ji ,&nbsp;Peng Song ,&nbsp;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}

{"title":"Research on the multi-signal DOA estimation based on ResNet with the attention module combined with beamforming (RAB-DOA)","authors":"Long Wu ,&nbsp;Yue Fu ,&nbsp;Xu Yang ,&nbsp;Lu Xu ,&nbsp;Shuyu Chen ,&nbsp;Yong Zhang ,&nbsp;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}

{"title":"Prior knowledge-guided multi-scale acoustic metamaterial sensing for gearbox weak fault signal detection","authors":"Yaqin Wang ,&nbsp;Jia Liu ,&nbsp;Huafei Pan ,&nbsp;Zhao Huang ,&nbsp;Jiaowei Xiao ,&nbsp;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}

{"title":"Design of subwavelength wide bandwidth sound absorbers by inverse convolutional neural networks","authors":"Peter Hawes,&nbsp;Marco Boccaccio,&nbsp;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}

{"title":"Experimental analysis of metamaterial with improved high sound levels absorption using complex frequency plane","authors":"Gauthier Bezançon ,&nbsp;Olivier Doutres ,&nbsp;Olga Umnova ,&nbsp;Philippe Leclaire ,&nbsp;Thomas Dupont","doi":"10.1016/j.apacoust.2025.110529","DOIUrl":"10.1016/j.apacoust.2025.110529","url":null,"abstract":"<div><div>This study proposes using the complex frequency plane representation as a tool to quantify loss levels of a metamaterial at low sound levels, enabling the prediction of trends in absorption coefficient changes at high sound levels. A multi-resonant metamaterial composed of a series of thin annular cavities connected by a central perforation is considered which has been previously studied in the linear regime. With the analytical model developed for the linear regime, the representation of the complex frequency plane allows understanding whether a low value of absorption peak is due to excessive losses or, instead, to a lack of losses in the material. As sound level increase, material losses rise, leading to decrease in absorption peaks for structures with excessive losses and increase of peak absorption coefficient values for those with insufficient losses. Multi-resonant metamaterials with a constant main pore profile are selected to exhibit resonances with various loss levels, and measurements in a high sound level impedance tube are conducted to validate the expected changes in absorption coefficient. After that, an acoustic black hole is considered and a structure with two low frequency absorption peaks increasing with sound level and presenting a broad absorption band with low sensitivity to high sound levels is identified. The predictions are validated experimentally.</div></div>","PeriodicalId":55506,"journal":{"name":"Applied Acoustics","volume":"231 ","pages":"Article 110529"},"PeriodicalIF":3.4,"publicationDate":"2025-01-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143095922","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A method for controlling dipole characteristic noise sources inside duct based on acoustic shortcut channel","authors":"Zhenfei Zang,&nbsp;Xiaochen Zhao,&nbsp;Shiqi Zhang,&nbsp;Jie Guo","doi":"10.1016/j.apacoust.2025.110542","DOIUrl":"10.1016/j.apacoust.2025.110542","url":null,"abstract":"<div><div>This study proposes a passive noise control method using coupled loudspeakers to counteract dipole sound sources within ducts. Coupled loudspeakers create an acoustic shortcut channel in the duct, which cancels out the out-of-phase noise radiated upstream and downstream by dipole sources placed in the duct. This paper establishes an acoustic model of the coupling system based on plane wave theory and analyzes its attenuation effect on duct noise. Additionally, the study analyzes the influence of geometric and loudspeaker parameters. To broaden the noise reduction frequency band, the paper investigates the acoustic attenuation characteristics of arrays of coupled loudspeakers combinations. Ultimately, experimental results demonstrate that properly arranged coupled loudspeakers can achieve more than 10 dB noise control effectiveness in the low-frequency range against dipole sources.</div></div>","PeriodicalId":55506,"journal":{"name":"Applied Acoustics","volume":"231 ","pages":"Article 110542"},"PeriodicalIF":3.4,"publicationDate":"2025-01-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143095924","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}

{"title":"Estimation of vibro-acoustic boundary conditions based on an experimental acoustic two-port characterization of a flexible flow duct","authors":"Jurgen Kersschot ,&nbsp;Hervé Denayer ,&nbsp;Wim De Roeck ,&nbsp;Wim Desmet","doi":"10.1016/j.apacoust.2025.110544","DOIUrl":"10.1016/j.apacoust.2025.110544","url":null,"abstract":"<div><div>For the accurate numerical prediction of the vibro-acoustic interactions in flow ducts, correct structural boundary conditions modeling the joints of flexible duct walls are key in the low frequency region. Therefore, a more elaborated model than the standard simply-supported or clamped conditions is needed. Robin boundary conditions allow a better approximation, although they introduce additional parameters of which the concrete values are unknown. In this paper, an indirect methodology is proposed to estimate these parameters, using two-port data obtained from acoustic measurements as reference for iterative optimization. As the two-port method is independent of the acoustic boundary conditions at the duct in- and outlet, the indirect methodology is useful for early-design phases when the whole duct system is unknown. To illustrate the proposed methodology, a flexible duct segment is tested and modeled with the finite element method. The structural boundary conditions are parametrized with frequency-independent stiffness and damping coefficients for linear spring-damper combinations. Non-uniform, mesh-independent distributions along the boundary are used and the pretension imposed by the joint is included. Optimization leads to parameter values improving the match with the reference data compared to standard boundary conditions. An experimental Operational Modal Analysis validates that the optimized numerical model not only approximates well the two-port data, but also captures the underlying vibro-acoustic physics, making the updated model valuable for vibro-acoustic predictions. To demonstrate the robustness of the methodology, a parameter study is conducted with numerical reference data, which shows that the structural boundary estimation is independent of structural material properties.</div></div>","PeriodicalId":55506,"journal":{"name":"Applied Acoustics","volume":"231 ","pages":"Article 110544"},"PeriodicalIF":3.4,"publicationDate":"2025-01-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143095926","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}

{"title":"A universal empirical model for materials in acoustic application","authors":"R.K. Dunne","doi":"10.1016/j.apacoust.2025.110549","DOIUrl":"10.1016/j.apacoust.2025.110549","url":null,"abstract":"<div><div>Sound absorption materials are essential for noise reduction across various industries, yet existing empirical models often lack accuracy, especially for low-density and thin materials. This study introduces two innovative models to address these limitations: an optimized Delany-Bazley model and a novel dimensionless parameter model derived using Buckingham’s Π-theorem. The optimized Delany-Bazley model refines its coefficients using a heterogeneous dataset of synthetic, plant-based, and animal fibres, achieving a 48.64% reduction in prediction error compared to the original model. The dimensionless parameter model eliminates the need for airflow resistivity and experimental testing by leveraging material properties such as fibre diameter, thickness, bulk density, and fibre density, resulting in a 44.92% error reduction. Both models demonstrate exceptional predictive accuracy within the realm of empirical modeling, with Noise Reduction Coefficient (NRC) errors as low as 2.92% for foam absorbers and even lower for certain fibrous samples. Furthermore, a novel NRC graph was developed, enabling rapid acoustic performance estimation based solely on material properties. These advancements improve the applicability of empirical models, offering versatile tools for Finite Element Analysis (FEA) simulations and sustainable noise control solutions. By addressing the limitations of existing empirical models, these innovations provide more accessible and accurate methods for predicting sound absorption across diverse materials, particularly those that are thin, low-density, and/or natural fibre-based.</div></div>","PeriodicalId":55506,"journal":{"name":"Applied Acoustics","volume":"231 ","pages":"Article 110549"},"PeriodicalIF":3.4,"publicationDate":"2025-01-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143095927","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Low bit-rate speech coding with predictive multi-level vector quantization","authors":"Xingye Yu,&nbsp;Ye Li,&nbsp;Peng Zhang,&nbsp;Lingxia Lin,&nbsp;Tianyu Cai","doi":"10.1016/j.apacoust.2025.110538","DOIUrl":"10.1016/j.apacoust.2025.110538","url":null,"abstract":"<div><div>During the development of modern communication technology, although wideband speech coding can provide high-fidelity speech transmission, its high bandwidth requirements limit its application in resource-constrained environments. Narrowband speech coding still holds research value. However, traditional narrowband low bit-rate speech coding methods usually cannot generate satisfactory speech quality. To address this issue, this paper proposes a narrowband low bit-rate speech coding architecture called PMVQCodec, with the following major improvements. Firstly, we design a predictive multi-level vector quantization (PMVQ) technique, which employs a predictor to effectively capture the correlations between latent frame vectors and combines it with multi-level vector quantization to enhance quantization efficiency. Additionally, we also introduce a full-band feature extractor to effectively reduce the computational complexity. In our experiments, both subjective and objective evaluations demonstrated the effectiveness of the proposed PMVQCodec architecture. Our proposed method can achieve higher quality reconstructed speech than Encodec and HiFiCodec at 1.2 kbps and 2.4 kbps, and even outperforms LyraV2 at 6 kbps.</div></div>","PeriodicalId":55506,"journal":{"name":"Applied Acoustics","volume":"231 ","pages":"Article 110538"},"PeriodicalIF":3.4,"publicationDate":"2025-01-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143095928","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}

{"title":"Enhancing underwater single snapshot DOA estimation for limited dataset with modified knowledge distillation","authors":"Qinzheng Zhang ,&nbsp;Haiyan Wang ,&nbsp;Xiaohong Shen ,&nbsp;Yongsheng Yan ,&nbsp;Yingying Zhu ,&nbsp;Jesper Rindom Jensen","doi":"10.1016/j.apacoust.2025.110531","DOIUrl":"10.1016/j.apacoust.2025.110531","url":null,"abstract":"<div><div>In recent years, the progress in DOA estimation using deep learning algorithms has attracted significant attention. However, their heavy reliance on extensive datasets poses a critical limitation, particularly in underwater settings where data collection is arduous. Furthermore, the lack of temporal correlation and statistical properties inherent in single-snapshot information lead to low accuracy in single-snapshot DOA estimation. To confront the above hurdles, this paper introduces an approach to improve the accuracy of underwater single snapshot DOA estimation with limited underwater datasets. By modifying the process structure and model characteristics of knowledge distillation (KD), we construct a new distillation structure that can bridge the gap between single snapshot data and multi-snapshot data sharing identical labels, achieving a breakthrough in compressing multi-snapshot data. This enhances the neural network's capacity to process both few-snapshot datasets and single-snapshot datasets. In addition, we designed novel input features to reduce the difficulty of CNN fitting by extracting the real and imaginary parts of the analytical signals, and integrated the array structure information to improve the generalization ability of our network in different scenarios. Besides, based on these innovations, we build a mapping framework between synthetic and real underwater datasets. This work involves second-order joint training of KD and transfer learning, which can help deal with small samples. The experiment results of our method show significant improvements in underwater DOA estimation accuracy, coupled with a marked reduction in overfitting risks associated with limited datasets. This work not only advances the application of deep learning in challenging underwater scenarios but also lays a foundation for future data-driven inference strategies.</div></div>","PeriodicalId":55506,"journal":{"name":"Applied Acoustics","volume":"231 ","pages":"Article 110531"},"PeriodicalIF":3.4,"publicationDate":"2025-01-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143095421","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}