运用神经网络推算音乐厅满场混响时间

混响时间是厅堂音质评价中的一个重要客观参量。本文提出了运用神经网络由空场混响时间推算满场混响时间的新方法。与其它方法比较,该方法有相当的预测精度。在此基础上,本文进一步分析了音乐厅中座椅类型和座椅数量对满场混响时间的影响,并对满场条件下座椅的吸声增量进行了讨论。     

音乐厅音质设计中响度评价参量的讨论

强感级(G,dB)是评价音乐厅音质设计中重要客观参量之一,它由厅内声级按标准声源的声功率级归一化得出。我们不仅考察厅内的总强感级G,还得分别考虑其早期声和后期声的G_早和G_后,因为它们的成因和主观效果是不同的。近年从主观试验结果中获得一个意外发现,即最低频段的G_低值与中频G_中值之比,将是决定厅内低音感的主要因素,而不是过去所说的低频和中频混响时间之比。10年前,人们认为环绕感LEV只由到达听者耳际的侧向声能所决定,如今则了解到后期总声能G_后这一因素也很重要。并提出由G和侧向比值LF来估算LFV的新公式,便于设计考虑。本文对音乐厅最佳G值作了讨论。近年的一个重要发现是,音乐厅内听者响度判断主要由混响声决定,或认为因视听综合心理效应使听者的响度判断与离舞台距离几乎无关,或两者兼具。后者涉及视觉信息输入影响听觉响度体验,即所谓多维感知综合效果的心理学问题。近年有人认为G参量不能直接反映音乐厅内听众听到的绝对声压级为由,提出以音乐片段中乐队齐奏强音标志段声级L_pf作为评价音乐厅响度的新物理指标。作者对此作了全面评析,说明其存在问题和不可行性。     

Nonexponential sound decay in concert halls

The paper presents acoustic measurement results for two concert halls in which nonexponential sound decay is observed. Quantitative estimates are given for how the obtained decay laws differ from exponential. Problems are discussed that arise when using reverberation time to assess the quality of room acoustics with nonexponential sound decay.     

Estimating the mixing time of concert halls using the eXtensible Fourier Transform

In room acoustics, we measure room impulse responses (RIRs) in order to fully describe the hall. RIRs are composed of a succession of arrivals (i.e., some sound rays which have undergone one or more reflections on their way from the source to the receiver). We propose the eXtensible Fourier Transform (XFT) in order to investigate the time evolution of spectral components of RIRs. The phase evolution versus time allows to estimate the mixing time, which is defined as the time it takes for initially adjacent sound rays to spread uniformly across the room. After presenting some properties of the XFT, we show why one must compensate the natural energy decay of the RIR in order to obtain stationary signals. We estimate mixing times for a set of experimental RIRs and several that are synthesized using a stochastic model. Then, we estimate the dependance of mixing time upon the source/receiver distance in all these RIRs. Results are consistent up to the lack of reproducibility of the sound sources, but are strongly dependent on some parameters used for computing the XFT. We finally discuss the relevance of the name mixing time with respect to the theory and in regard to the time we estimate, that we propose to call cross-over time.     

Audience noise in concert halls during musical performances

Noise generated by the audience during musical performances is audible and sometimes disturbing. In this study, an attempt to estimate such audience noise was carried out. From the recordings of performances in five performance spaces (four concert halls and one opera house), probability density functions of the sound pressure levels were obtained in octave bands, which were fitted with three Gaussian distribution curves. The Gaussian distribution curve with the lowest mean value corresponds to a mixture of the technical background noise and audience generated noise, which is named the mixed background noise. Finally, the audience noise distribution is extracted by energy subtraction of the technical background noise levels measured in an empty condition from the mixed background noise levels. As a single index, L(90) of the audience noise distribution is named the audience noise level. Empirical prediction models were made using the four orchestra concert halls, revealing that the audience noise level is significantly correlated with the technical background noise level. It is therefore concluded that a relaxation of the current background noise recommendations for concert halls is not recommended.     

Monaural and binaural parameters of Rudolfinum concert halls in Prague

Objective parameters for the evaluation of the Rudolfinum concert hall in Prague, Czech Republic are the focus of the present article. The measured results for Reverberation parameters, Energy parameters, Intelligibility parameters, and Spatial parameters of the building’s two halls are presented and discussed including a comparison with recommended values or theory, as well as several unique architectural and acoustical qualities of the halls. The early lateral energy fraction parameter is measured by the intensity probe method discussed in the supplement. The performance is verified by tests in anechoic and reverberant rooms.     

Distribution of selected monaural acoustical parameters in concert halls

Acoustical parameters calculated from impulse responses are often used to evaluate the characteristics of concert halls. Representative parameters are listed in the Annex of ISO 3382 and the methods of calculation and the minimum number of measurement positions are explained in detail. However, a method for selecting measurement positions is not discussed clearly, because there are wide variations in the characteristics of sound fields. This report provides basic data to solve this problem using spatial distribution characteristics of parameters in halls. Three large-scale measurement campaigns were conducted in which impulse responses were measured at 1427, 180, and 511 locations. Relatively large differences in the obtained parameters compared with well-known difference limens suggest that determining the distributions of parameter values is important. Contour maps are therefore used to display the distributions along with mean values.     

Lateral reflections are favorable in concert halls due to binaural loudness

A recent study on perceptual difference in simulated concert halls showed that a concert hall renders stronger sound with more bass when the temporal envelope of a signal is preserved in the reflections [Lokki et al., J. Acoust. Soc. Am. 129, EL223-EL228 (2011)]. In the same study the lateral reflections were shown to contribute to the perceived envelopment and openness. Moreover, the listening test results suggest that lateral reflections contribute to perception of sound source distance. Here, it is shown that lateral reflections are beneficial due to their increasing effect on binaural loudness-the phenomenon known well in psychoacoustics, but not in architectural acoustics. The reflections from the side are amplified more than median plane reflections, in particular at high frequencies, due to the shape of the human head.     

座位数在音乐厅和歌剧院计算中的作用

在对Leo Beranek著《音乐厅和歌剧院》一书中所提供的80余所音乐厅和歌剧院声学参数的测定结果数据分析的基础上,经过简单的数学处理和简化,提出一些个人见解:强度因子G和座位数N的关系;对于N小于和等于2000的音乐厅,空场早期衰变时间EDT等于空场混响时间。文中给出在N＜ 2000的情形下空场EDT的计算。座椅数是决定厅堂容积的出发点,由此做出两个厅堂的初步设计实例。     

On the combined effects of early- and late-arriving sound on spatial impression in concert halls

This paper describes six new experiments involving subjective ratings of the listener envelopment, LEV, and the apparent source width, ASW, of simulated sound fields. Previous work has identified LEV and ASW as the principal components of spatial impression in concert halls and has shown that ASW is primarily influenced by the level of early lateral reflections and LEV by late-arriving lateral reflections. The new results in this paper show that LEV can result from nonlateral late-arriving sounds and demonstrate the conflicting effects of early- and late-arriving lateral sound on ASW and LEV when both are present, as would occur in real halls. While it is possible to create simulated sound fields with only either LEV or ASW, in typical concert halls, the balance between early- and late-arriving lateral sound will determine the relative importance of LEV and ASW. LEV and ASW are shown to be perceived when the critical components of the sound field are salient relative to other components. The results of the new subjective studies were used to estimate expected ASW and LEV in 16 halls. In these halls LEV is predicted to be the stronger component of spatial impression.     

Design of room acoustics and a MCR reverberation system for Bjergsted concert hall in Stavanger

Bjergsted concert hall in Stavanger, Norway, was opened in 1982 as a multi-purpose hall, but primarily intended for concerts. The hall has been rebuilt from an old dome-shaped exhibition hall. The new shape of the hall was designed using computerized sound ray tracing, in a three-dimensional mathematical model. The hall is fan-shaped and additional lateral reflections are provided by large reflectors hanging freely below the ceiling. Because of the marginal volume per seat and the hall's many light-weight constructions, we have designed, specifically for this hall, an electroacoustic multi-channel reverberation system. This increases the reverberation time over a broad frequency range and, in addition, the sound level is also increased.     

Subjective preference tests for sound fields in concert halls simulated by the aid of a computer

In the context of optimal acoustic design of concert halls, preference tests for simulated sound fields with various combinations of early discrete reflections and subsequent reverberation were conducted and the results compared with objective parameters. The results show that lateral early reflections which yield a small value of the interaural cross correlation give more preferred judgments than do non-lateral ones. The most preferred initial time delay (or scale factor for the dimensions of concert halls) can be determined by the long-time autocorrelation function of a music motif and by the total amplitude of both early reflections and the subsequent reverberation. As a comprehensive design concept, an “overall preference” for listeners is introduced here from the preference results.     

Loudness perception of low tones undergoing partial masking by higher tones in orchestral music in concert halls

Objective acoustical parameters for halls are often measured in 1-octave bands with mid-frequencies from 125 to 4000 Hz. In reality, the frequency range of musical instruments is much wider than that, and the fundamentals of the lower notes of bass instruments are contained in 31.5 or 63?Hz bands. Overtones of fundamentals in these bands fall in 125?Hz band. This report presents subjective experiments designed to determine to what extent the overtones in 125?Hz band and higher bands influence the loudness sensation of the components in 63?Hz band. In the experiments, the 125?Hz and higher components of the musical tone are used to act as a masker against the lower component used as a maskee. The threshold of the difference between G(125?Hz) and G(lower band) that just enables one to hear the fundamental tones in the lower band is determined. Masked loudness of 63?Hz sinusoidal tone caused by partial masking noise with higher frequencies was determined based on a similar procedure to the masked loudness-matching function. The result indicates that the difference in loudness of low tone will not be noticeable even if G changed by ±2.5 to ±3?dB, at least when there are other accompanying instruments.     

Relation of acoustical parameters with and without audiences in concert halls and a simple method for simulating the occupied state

Five acoustical parameters-reverberation time RT, early decay time EDT, clarity C80, strength G, and interaural cross-correlation coefficient IACC-were measured using identical procedures with and without audiences in six concert and opera halls. Reverberation times without audiences were measured in 15 additional halls using the same measuring techniques as for the six halls above, but for full occupancy the data were taken from musical stop chords at symphonic concerts. This paper shows that in all halls (1) the occupied RT can be predicted from the unoccupied RT using a linear regression equation, y = a - b exp(x), within acceptable limits, at low- and mid-frequencies. It is also shown for the six halls that (2) occupied C80's are predicted accurately from unoccupied values by the newly proposed equation; (3) G's with and without audiences are highly correlated by a first degree linear regression equation; and (4) IACCs have nearly the same value in both occupied and unoccupied halls. As a separate subject, the successful use of a cloth covering for seats in a concert or opera hall to simulate the occupied condition has been developed.     

The reverberation time of furnished rooms in dwellings

The study gives the results of the measurements of the reverberation time in 11,687 rooms, of which 11,457 are furnished (8246 bedrooms, 3211 living rooms) and 230 unfurnished. All the rooms have heavy walls and ceilings, and a heavy floor covering. The reverberation times measured are quite similar in bedrooms and living rooms within the same size range, and decrease fairly uniformly as the frequency increases. Moreover, in each frequency band the greater the volume of the room, the greater the reverberation time.The results of this extensive fieldwork allow us to predict accurately the reverberation time in these kinds of spaces as a function of their size and the frequency. These data may be useful for improving the accuracy of calculations models to estimate the reverberation time of enclosed spaces. A comparison between the reverberation time measured in this work and that proposed in PrEN ISO 10052.2002(E) has been made.     

Blind estimation of reverberation time in classrooms and hospital wards

This paper investigates blind Reverberation Time (RT) estimation in occupied classrooms and hospital wards. Measurements are usually made while these spaces are unoccupied for logistical reasons. However, occupancy can have a significant impact on the rate of reverberant decay. Recent work has developed a Maximum Likelihood Estimation (MLE) method which utilises only passively recorded speech and music signals, this enables measurements to be made while the room is in use. In this paper the MLE method is applied to recordings made in classrooms during lessons. Classroom occupancy levels differ for each lesson, therefore a model is developed using blind estimates to predict the RT for any occupancy level to within ±0.07 s for the mid-frequency octave bands. The model is also able to predict the effective room and per person absorption area.Ambient sound recordings were also carried out in a number of rooms in two hospitals for a week. Hospital measurements are more challenging as the occurrence of free reverberant decay is rarer than in schools and the acoustic conditions may be non-stationary. However, by gaining recordings over a period of a week, estimates can be gained within ±0.07 s. These estimates are representative of the times when the room contains the highest acoustic absorption. In other words when curtains are drawn, there are many visitors or perhaps a window may be open.