听觉感知中的噪声特性语义描述及其分析

基于听觉感知的噪声语义描述是噪声声品质研究的基础性问题,已有研究未将语义描述与噪声来源、频谱特性以及产品运行状态等物理信息联系起来。该文分别针对飞机舱内噪声、车辆噪声和空气净化器噪声这3组典型噪声开展了主观评价实验,并通过多维尺度分析和主成分分析描述了3组噪声的语义空间,系统分析了不同类型噪声的描述词,同时解释了描述词与噪声物理属性之间的联系。研究发现：飞机舱内噪声、车辆噪声以及空气净化器噪声可以由4维、4维和3维语义空间进行描述;不同类型噪声在语义描述中具有共性与个性,3组噪声语义的主要维度均与嘈杂感相关,而噪声的个性描述词与其声源的物理属性密切相关;进行声品质建模及应用时,应同时考虑噪声共性和个性描述词对听觉感知的影响,采取有针对性的措施以提升产品声品质。该文从听觉感知的角度进行了噪声特性的语义描述和分析,研究结果可为产品声品质以及噪声控制研究提供帮助。     

采用生理响度感知机制的车内噪声声品质预测

现有车内噪声声品质预测的响度计算中没有考虑真实人耳生理解剖结构的传声特性,因此提出了一种基于生理响度感知模型的车内噪声声品质预估方法。首先,采集两款轿车的车内噪声样本,并通过主观评价试验得到车内噪声的主观评价烦恼度。之后,整合中耳集总参数模型与耳蜗传输线模型,构建生理响度模型。然后,以生理响度模型的响度计算值为主要参数,结合尖锐度、粗糙度与车内噪声的主观评价值,通过TabNet模型构建了车辆声品质预测模型。最终,对比分析了所构建声品质模型与基于现有标准响度模型所构建的声品质模型的预测效果。结果表明,采用生理响度模型的声品质预测平均误差百分比仅有4.73%,优于采用Moore响度(6.13%)与Zwicker响度(6.94%)的声品质预测结果。此外,所构建的TabNet声品质预测模型的平均误差百分比也低于基于BP神经网络模型的平均误差百分比(7.60%)。采用生理响度模型的TabNet声品质预测能够提高车内噪声声品质客观评价的准确率。     

典型城市地铁站台噪声满意度*

地铁是城市轨道交通重要组成部分,地铁站台噪声对乘客身心健康有重要影响。以合肥轨道交通一号线为例,具体选择地下侧式、地下双岛式和地下岛式3个典型空间类型的地铁站台为研究对象,通过噪声测量和问卷调查相结合的方法,针对地铁在不同运营时段的乘客站台噪声满意度与噪声声元素舒适度进行评价,探讨性别、声元素舒适度以及站台噪声对乘客噪声满意度的影响。结果表明：对地铁站台噪声满意度评价,乘客性别差异不显著;在乘车高峰期时段,地铁站台广播提示声、地铁工作人员吹哨声、列车进站声和列车出站声是影响乘客站台噪声满意度主观评价的主要噪声声元素;乘客噪声满意度主观评价与站台噪声值呈强负相关,噪声值越高,满意度越低;乘客噪声满意度为可接受时的地铁站台噪声阈值,在乘车高峰期时段为74 dB(A),非高峰期时段为67 dB(A)。该文工作可为城市地铁站台噪声满意度评价及其空间结构设计提供科学参考。     

混合算法优化的乘用车舱内噪声感知烦恼度神经网络建模

针对传统乘用车舱内噪声感知烦恼度量化模型精度低的问题,提出了一种利用混合算法优化的神经网络模型预测舱内噪声感知烦恼度的评价方法。此混合算法融合麻雀搜索算法(SSA)和遗传算法(GA),对反向传播(BP)神经网络进行优化,根据声品质主客观评价数据,建立SSA-GA-BP网络的乘用车舱内噪声感知烦恼度客观量化模型,与BP模型、GA-BP模型、SSA-BP模型进行对比分析。结果表明, SSA-GA-BP模型能够实现更高的预测精度,更接近主观评价数值,泛化能力更强,可替代传统的声品质主观评价实验。     

滨海公共空间声环境感知评价及影响因素*

滨海公共空间是滨海城市人居环境的重要组成部分,其声环境质量是影响公共空间环境体验的关键因素。本研究以青岛沿海岸线四类滨海公共空间（步道类、广场类、公园类、沙滩类）的声环境为对象,通过问卷调查和实地测量探索滨海公共空间声环境的特征、感知评价及其影响因素。结果表明：（1）滨海公共空间当前环境噪声水平基本不超过国家标准限值55dB(A) ,各类空间声环境客观特征显著不同：公园噪声水平最低,自然声源丰富;广场噪声水平最高,各种城市噪声较多;步道噪声水平较低,但受到交通噪声的影响;沙滩噪声水平较高,但以中低频的海浪声为主,比较稳定。（2）各类滨海公共空间的声源感知以自然声为主,但是广场的自然声源感知较少。此外,声环境的愉悦感评价普遍较高,但在事件感、吵闹度和总体满意度评价上具有显著差异：沙滩声环境总体评价最好,其次是公园,广场和步道的声环境则亟待改善。（3）交通噪声和最大声级Lmax对声环境的负面影响最为显著,此外受访者年龄也是影响滨海公共空间声环境感知评价的关键因素。研究结论可以为滨海公共空间声环境的优化设计提供理论依据。     

基于非度量多维尺度分析的噪声声品质主观评价研究

以等响处理后双耳记录的车内噪声为例,针对非度量多维尺度分析(NMDS)是否适用于研究一致性较强的噪声声品质主观评价这一疑问进行研究。研究发现,按照评价者之间正负相关性分类处理的方法,对于运用NMDS揭示了一致性较强的复杂车内噪声声品质潜在的主观感知结构起着重要作用。采用该方法可以揭示出:车内噪声的偏好性可以作为主观感知相似性主要子维度的表示;评价者分类处理之后,两类评价者对于车内噪声的偏好性均可采用一个维度表示,并可直接采用成对比较法的直接排序法获得;一类评价者的偏好性与低沉度成正比,而另一类评价者与低沉度成反比。     

汉语普通话的平均频谱

本文描述了用混响室中重迭的语言声来测定语言平均频谱的方法,用这个方法测出了汉语普通话的平均频谱.测量结果与自由声提语言噪声的测量结果对比,同时也与由各个元音频谱平均计算出的平均频谱比较,发现几种结果互相符合得很好.由此,给出了普通话的标准频谱以及它的解析表达式.     

空调压缩机噪声测试及其声品质客观参量分析*

针对空调压缩机辐射声场的不同场点噪声声品质的差异,采用心理学声品质参量即响度、尖锐度、总感觉噪度,对压缩机辐射声场的不同场点噪声进行声品质客观参量分析,研究压缩机场点噪声声品质客观参量值变化规律。通过特征响度研究各场点噪声响度在频域的分布,并得出压缩机各场点噪声特征响度峰值所在噪声频带。实验结果表明,对于处在压缩机不同方向上的场点,其噪声声品质客观参量值存在明显差异,但其噪声特征响度峰值却处在相同的噪声频带。研究工作为压缩机全方向降噪提供参考依据也为研究压缩机的噪声声品质探索了一种新思路。     

采用响应面回归的汽车多属性声品质预测方法*

单一主观评价分数无法准确描述人耳对声品质的多属性偏好特征。该文在多属性声品质试验数据的基础上,以愉悦度、平顺度和驾驶乐趣的主观评价分数为因变量,通过相关分析筛选出响度、尖锐度和A计权声压级3个主要自变量,引入响应面回归方法,分别建立因变量与自变量之间的预测模型,通过与多元线性模型和BP神经网络模型对比验证了精度。最后,建立多属性主观评价分数之间的量化映射模型。该研究可为多属性汽车声品质的优化控制提供参考。     

城市历史街区声景观及视听感知实验研究*

城市历史街区声景观是空间环境的重要内容,直接影响人们的感知体验。论文以济南老城泉水街区为例研究城市历史街区的声景观并进行视听感知实验。结果表明,现场声元素喜好度显著高于实验室单听评价。鸟鸣、鹅叫声在现场和实验室评分都较高,大流速流量水景声在现场评分高而实验室评分低,小流速流量水景声在实验室和安静环境中评分较高,在吵闹环境中评分较低;声环境满意度与视觉景观优美度、声源组成喜好度、安静度之间存在显著正相关关系,与环境声压级存在显著负相关关系,但当较高的环境声压级主要由水声贡献时,声环境满意度明显提高;应从提高景观优美度、优化声源组成、降低环境声压级等方面进行声景观设计。论文为声元素喜好度和声环境满意度研究提供参考。     

Comparison between old and new noise standards in Nagoya City

The Japanese Environmental Agency (now the Ministry of the Environment) updated the environmental quality standards for noise in April 1999. The new standards replaced the median value of percentile level L₅₀ for noise evaluation with the equivalent sound pressure level L_Aeq. The standards renewed the classification of areas and time sections. The most significant change was the introduction of category of artery-road-adjacent area.This report sets the range of the artery-road-adjacent area to 20 m or less from the applicable road to compare the new standards with the old, based on data collected in Nagoya City. The achieved rates for the new standards seem to be on the whole the same as those for the old standards. However, a detailed analysis reveals some differences, such as higher achieved rates in the artery-road-adjacent areas and lower achieved rates in the general areas for the new standards than for the old.     

Index for vehicle acoustical comfort inside a passenger car

Sound quality is among the main factors that influence customers’ preference for choosing good automobile products. It all started more than 10 years ago and grows up so fast due to high competition in the automotive industries. A-weighted noise levels and sound power are usually utilised to measure the noise but they are not adequate to characterise the impact sound inside a car. The most popular approach to determine sound quality of a product is to define an annoyance or specific index, which involves both subjective and objective evaluations. Subjective and objective tests should be studied concurrently in order to determine the sound quality inside a passenger car. This approach is used in this paper to evaluate vehicle comfort index according to most frequently used sound quality metrics, namely; Zwicker loudness, sharpness, roughness and fluctuation strength. As a result researchers of different fields of automotive acoustics investigations can use this index according to the type of road (international road roughness) without any need to perform time-consuming jury tests. The metrics are correlated with jury test results that show which of them and how much has affected the acoustical comfort of the vehicle. The relation between road roughness and vehicle acoustical comfort index is another point of interest in this research.     

Feature extraction for sound source material recognition with impact sounds

Sound source recognition is a part of environmental sound recognition,which is one of the most important research areas in pattern recognition.Impact sounds carry much physical information associated with the sound sources,which makes impact sound based sound source recognition an important approach to improve recognition performance.In this study,the impact sound continuum synthesized with a ball-plate collision model is used for material recognition of the impacted plates.The basis function learning method and time-frequency representation methods,including the short time Fourier transform and the wavelet packet transform,are applied into classification and the recognition results are compared.The result shows that the features obtained by using the basis function learning perform better for material classification of the impacted plates than that by using the short time Fourier transform and the wavelet packet transform.This demonstrates the high efficiency and superiority of this method in material recognition of sound sources.     

利用冲击声辨识声源材料的特征提取

声源辨识属于环境声识别的范畴,是模式识别的一个重要研究方向。冲击声携带了大量的声源物理信息,因此利用冲击声提取特征进行声源材料辨识是提高声目标识别分类性能的重要途径。对球-板撞击物理模型合成的冲击声连续统,提出使用基函数学习法提取目标特征,同时利用短时傅里叶变换和小波包变换进行特征提取,以此为基础完成被击平板的材料识别。研究结果表明,利用基函数学习法获得的特征,对于冲击声分类的效果明显优于短时傅里叶变换和小波包变换方法,说明利用该方法进行冲击声声源材料辨识的可行性和优势。      

Standardised acoustic characterisation of sonic crystals noise barriers: Sound insulation and reflection properties

This work presents sound insulation and sound reflection measurements conducted over sonic crystal noise barriers according to the European standards EN 1793-2, EN 1793-5 and EN 1793-6. In most of the reference literature, sound insulation and reflection properties of sonic crystals are measured or a diffuse sound field or in a direct sound field including the top and side edge diffraction effects together with the transmitted (or reflected) components. The aim of this work is to perform free-field measurements over a real-sized sample in order to window out all diffraction components and to verify the points of strength and weakness of the application of standardised measurements to sonic crystals. Diffuse field measurements in laboratory are also done for comparison purposes. Since the target frequency range for traffic noise spectrum is centred at around 1000 Hz, a finite element based parametric investigation is performed to design unit cells capable of generating band gaps in the one-third octave bands ranging from 800 Hz to 1250 Hz. Then, 3 × 3 m sonic crystal noise barriers are installed in the Laboratory of the University of Bologna and sound insulation and sound reflection measurements are performed according to the mentioned active standards for normal incidence. Sound insulation is measured for diffuse incidence too. The two methods give different results. The method more directly comparable to calculations is the free-field one. However, if on the one hand the application of a time window allows to compute the transmitted or reflected component only, on the other hand the time window itself limits the maximum width of the sample for which all reflections of the n-th order having a significant spectral content are included, and thus results critical in the analysis of this kind of noise barriers. Nevertheless, the standardised measurements allow a direct comparison between the performance of sonic crystals and common noise barriers.     

Evaluating airborne sound insulation in terms of speech intelligibility

This paper reports on an evaluation of ratings of the sound insulation of simulated walls in terms of the intelligibility of speech transmitted through the walls. Subjects listened to speech modified to simulate transmission through 20 different walls with a wide range of sound insulation ratings, with constant ambient noise. The subjects' mean speech intelligibility scores were compared with various physical measures to test the success of the measures as sound insulation ratings. The standard Sound Transmission Class (STC) and Weighted Sound Reduction Index ratings were only moderately successful predictors of intelligibility scores, and eliminating the 8 dB rule from STC led to very modest improvements. Various previously established speech intelligibility measures (e.g., Articulation Index or Speech Intelligibility Index) and measures derived from them, such as the Articulation Class, were all relatively strongly related to speech intelligibility scores. In general, measures that involved arithmetic averages or summations of decibel values over frequency bands important for speech were most strongly related to intelligibility scores. The two most accurate predictors of the intelligibility of transmitted speech were an arithmetic average transmission loss over the frequencies from 200 to 2.5 kHz and the addition of a new spectrum weighting term to R(w) that included frequencies from 400 to 2.5 kHz.     

Measurement of sound absorption coefficients of flat surfaces in a workshop

To improve the acoustic treatment of facings and provide appropriate solutions for noise control at workplace, it is necessary to develop methods of acoustic characterization of the walls in industrial halls. Sound absorption coefficient measurement in industrial rooms is however quite a difficult task because of the partially reverberant conditions. This work describes the measurement of the sound absorption coefficient of flat panels subject to small angle sound incidence, in an industrial hall using an experimental device equipped with an acoustic array. The directivity of this array has been optimized so that the major part of the received acoustic energy would come from one portion only of the investigated facing, this, in turn attenuating the reflected beams due to the reverberation. This new device includes an impulse sound source targeting the panels. The present article focuses mainly on the sound source design and implementation. It also describes some sound absorption measurements carried in a semi-anechoic chamber and in an industrial hall in order to examine the performance of the device. Sound absorption coefficients of several standard liners obtained through this device have been compared to those resulted from the two microphone technique.     

Study on the effect of sound duration on the annoyance of helicopter noise by applying a technique of time compression and expansion of sound signals

The number of helicopter operations has rapidly increased during the last 20 years in Japan. Helicopter noise sounds different from other aircraft; the waveform of the sound pressure is impulsive and the signal duration is relatively long. The Environmental Agency of Japan implemented new guidelines for evaluating noise exposure around small airports, including heliports, in 1990. This study was executed in connection with the development of provisional guidelines. Psychoacoustic experiments were carried out to identify an evaluation index for helicopter noise. In order to examine the effect of duration independently, we not only used original sound recordings, but also synthesized sound signals. The durations of these sound signals were time compressed or expanded without degrading the quality of the original sound recording. The test results show that the effect of duration is significant, and that the A-weighted sound exposure level is a better index than the maximum A-weighted sound pressure level for the evaluation of helicopter noise.