Concentration of reflected sound in a room treated with cube corner retroreflectorsa).

A room was treated to be predominantly retroreflective in the high frequency range by introducing arrays of cube corner retroreflectors (CCRs) over most surfaces (excluding the floor). In a small room (volume 55 m3), 156 CCRs in the form of square trihedra with 350 mm edge lengths were used as wall and ceiling treatment. The horizontal plane distribution of reflected energy was measured from omnidirectional sources, and a head and torso simulator was used to measure voice support. Results show a high concentration of reflected energy returned to omnidirectional source positions in high frequency octave bands (2-8 kHz). Finite-difference time-domain (FDTD) simulations of the room yielded similar distributions to the omnidirectional measurements, showing greater sound concentration when more CCRs are introduced. By contrast, FDTD simulation of an equivalent flat-surfaced room yielded no reflected sound concentration at the source, with results close to diffuse field theory in high frequency octave bands. Measured voice support values derived from oral-binaural room impulse responses exceed diffuse theory expectations by 5 dB. Thus, the paper demonstrates that retroreflective array treatment can change room acoustical conditions, concentrating reflected energy onto an arbitrarily located source.

Separate effects of auditory and visual room size on auditorium seat preference: a virtual reality study.

Audiovisual integrations and interactions happen everywhere, including in music concerts, where combined visual and auditory perception contributes to overall enjoyment. Thirty-three participants evaluated their overall subjective preference at various seats in four virtual auditoria, which comprised congruent and incongruent auditory and visual renders of two auditoria that differ only in size. Results show no significant difference between participants who completed the experiment in a fully calibrated and standardized laboratory environment and participants who completed remotely using various VR equipment in various environments. Both visual and auditory auditorium size have significant main effects, but no interaction. The larger hall is preferred for both conditions. Audiovisual congruency does not significantly affect preference.

Effect of voice support level and spectrum on conversational speech.

One's own voice (autophony) is transmitted to the ears as direct airborne sound, bone conduction, and indirect airborne sound from reflections characterized by overall gain and spectro-temporal features. This study investigates how the spectral profile and gain of simulated indirect airborne sound, quantified as voice support (STV), affect the speaking voice of talkers. Pairs of participants performed a conversation elicitation task in anechoic conditions. The indirect airborne sound was provided in real-time via open headphones that maintain the direct airborne transmission path. Experimental conditions included high-pass, low-pass, and all-pass versions of STV, each presented at three overall gains, and a Baseline condition with no electroacoustic contribution to STV. The results show an overall speech level reduction of 0.22 dB for every additional dB of speech-weighted STV, i.e., a -0.22 dB/dB slope. There was some effect of STV spectrum on speech: slope for the high-pass condition was steeper (statistically significant) and significantly different from the all-pass slope; spectral balance (2-4 kHz vs 0-2 kHz) of speech showed an interaction effect between gender and experimental conditions. This paper's findings may inform acoustic treatments in environments where overall sound reduction is of interest for favorable ergonomics and occupational health for voice professionals.

Chamber musicians' acoustic impressions of auditorium stages: Relation to spatial distribution of early reflections and other parameters.

Characterizing stage acoustics using objective parameters has seen some recent resurgence-several studies have noted the importance of the directionality of early stage reflections to musicians, which is not adequately represented using existing omnidirectional stage-support parameters. This study examines the subjective impressions of 19 chamber musicians against omnidirectional [reverberation time, early and late support (STEarly, STLate), etc.], and proposed spatially-defined parameters (TH and TS), along with simple ratios of stage dimensions derived from measurements on eight purpose-built stages. TH is a ratio of early energy from "above" to that from the "horizontal," while TS relates energy from above to that from the "sides" of the stage. Robust mixed-effects analyses showed that the musicians' overall acoustic impression ratings are predicted (i) by TH within a linear model; (ii) by TH × STEarly,TH × STLate, and TS × STEarly,TS × STLate; (iii) by STEarly, STLate each within parabolic models; and (iv) by several architectural parameters' linear and parabolic models. These findings reinforce recent studies of spatially-defined parameters to more fully account for the subtleties of onstage sound fields. Some simple design recommendations are presented, although future studies are needed to confirm these findings/recommendations for a wider range of auditorium stages.

Effect of a chamber orchestra on direct sound and early reflections for performers on stage: A boundary element method study.

Early reflections are known to be important to musicians performing on stage, but acoustic measurements are usually made on empty stages. This work investigates how a chamber orchestra setup on stage affects early reflections from the stage enclosure. A boundary element method (BEM) model of a chamber orchestra is validated against full scale measurements with seated and standing subjects in an anechoic chamber and against auditorium measurements, demonstrating that the BEM simulation gives realistic results. Using the validated BEM model, an investigation of how a chamber orchestra attenuates and scatters both the direct sound and the first-order reflections is presented for two different sized "shoe-box" stage enclosures. The first-order reflections from the stage are investigated individually: at and above the 250 Hz band, horizontal reflections from stage walls are attenuated to varying degrees, while the ceiling reflection is relatively unaffected. Considering the overall effect of the chamber orchestra on the direct sound and first-order reflections, differences of 2-5 dB occur in the 1000 Hz octave band when the ceiling reflection is excluded (slightly reduced when including the unobstructed ceiling reflection). A tilted side wall case showed the orchestra has a reduced effect with a small elevation of the lateral reflections.

Comparison of psychoacoustic-based reverberance parameters.

This study compared psychoacoustic reverberance parameters to each other, as well as to reverberation time (RT) and early decay time (EDT) under various acoustic conditions. The psychoacoustic parameters were loudness-based RT (TN), loudness-based EDT [EDTN; Lee, Cabrera, and Martens, J. Acoust. Soc. Am. 131, 1194-1205 (2012a)], and parameter for reverberance [PREV; van Dorp Schuitman, de Vries, and Lindau., J. Acoust. Soc. Am. 133, 1572-1585 (2013)]. For the comparisons, a wide range of sound pressure levels (SPLs) from 20 dB to 100 dB and RTs from 0.5 s to 5.0 s were evaluated, and two sets of subjective data from the previous studies were used for the cross-validation and comparison. Results of the comparisons show that the psychoacoustic reverberance parameters provided better matches to reverberance than RT and EDT; however, the performance of these psychoacoustic reverberance parameters varied with the SPL range, the type of audio sample, and the reverberation conditions. This study reveals that PREV is the most relevant for estimating a relative change in reverberance between samples when the SPL range is small, while EDTN is useful in estimating the absolute reverberance. This study also suggests the use of PREV and EDTN for speech and music samples, respectively.

The effect of loudness on the reverberance of music: reverberance prediction using loudness models.

This study examines the auditory attribute that describes the perceived amount of reverberation, known as "reverberance." Listening experiments were performed using two signals commonly heard in auditoria: excerpts of orchestral music and western classical singing. Listeners adjusted the decay rate of room impulse responses prior to convolution with these signals, so as to match the reverberance of each stimulus to that of a reference stimulus. The analysis examines the hypothesis that reverberance is related to the loudness decay rate of the underlying room impulse response. This hypothesis is tested using computational models of time varying or dynamic loudness, from which parameters analogous to conventional reverberation parameters (early decay time and reverberation time) are derived. The results show that listening level significantly affects reverberance, and that the loudness-based parameters outperform related conventional parameters. Results support the proposed relationship between reverberance and the computationally predicted loudness decay function of sound in rooms.

Non-auditory factors affecting urban soundscape evaluation.

The aim of this study is to characterize urban spaces, which combine landscape, acoustics, and lighting, and to investigate people's perceptions of urban soundscapes through quantitative and qualitative analyses. A general questionnaire survey and soundwalk were performed to investigate soundscape perception in urban spaces. Non-auditory factors (visual image, day lighting, and olfactory perceptions), as well as acoustic comfort, were selected as the main contexts that affect soundscape perception, and context preferences and overall impressions were evaluated using an 11-point numerical scale. For qualitative analysis, a semantic differential test was performed in the form of a social survey, and subjects were also asked to describe their impressions during a soundwalk. The results showed that urban soundscapes can be characterized by soundmarks, and soundscape perceptions are dominated by acoustic comfort, visual images, and day lighting, whereas reverberance in urban spaces does not yield consistent preference judgments. It is posited that the subjective evaluation of reverberance can be replaced by physical measurements. The categories extracted from the qualitative analysis revealed that spatial impressions such as openness and density emerged as some of the contexts of soundscape perception.

The effect of visual and auditory cues on seat preference in an opera theater.

Opera performance conveys both visual and auditory information to an audience, and so opera theaters should be evaluated in both domains. This study investigates the effect of static visual and auditory cues on seat preference in an opera theater. Acoustical parameters were measured and visibility was analyzed for nine seats. Subjective assessments for visual-only, auditory-only, and auditory-visual preferences for these seat positions were made through paired-comparison tests. In the cases of visual-only and auditory-only subjective evaluations, preference judgment tests on a rating scale were also employed. Visual stimuli were based on still photographs, and auditory stimuli were based on binaural impulse responses convolved with a solo tenor recording. For the visual-only experiment, preference is predicted well by measurements taken related to the angle of seats from the theater midline at the center of the stage, the size of the photographed stage view, the visual obstruction, and the distance from the stage. Sound pressure level was the dominant predictor of auditory preference in the auditory-only experiment. In the cross-modal experiments, both auditory and visual preferences were shown to contribute to overall impression, but auditory cues were more influential than the static visual cues. The results show that both a positive visual-only or a positive auditory-only evaluations positively contribute to the assessments of seat quality.

Influence of fundamental frequency and source elevation on the vertical localization of complex tones and complex tone pairs.

This study investigates the vertical localization of single complex tones (monads) and simultaneous complex tone pairs (dyads), especially as it is affected by their fundamental frequency and source elevation. Two complex tone timbres are considered: one consisting of five low-order harmonics, and the other of all odd harmonics (a square wave). Sound sources were at -15, 0, 15, and 30 deg from the horizontal plane at ear height. For eight subjects, this source array was in the median plane, and for a further nine subjects, it was directly to the subject's left (lateral plane). The subjects localized the angle of the auditory image(s) of one or two complex tones around the vertical plane containing the sound sources. Mean responses for the five-harmonic complex tones show a systematic effect (referred to as Pratt's effect) of fundamental frequency on vertical localization--whereby high-frequency complex tones are localized to positions higher than low-frequency complex tones for equivalent source positions. For the square wave, the sound-source position dominates localization, although some effect of fundamental frequency is evident for median plane sources.

The acoustic characteristics of professional opera singers performing in chorus versus solo mode.

In this study, members of a professional opera chorus were recorded using close microphones, while singing in both choral and solo modes. The analysis included computation of long-term average spectra (LTAS) for the two song sections performed and calculation of singing power ratio (SPR) and energy ratio (ER), which provide an indication of the relative energy in the singer's formant region. Vibrato rate and extent were determined from two matched vowels, and SPR and ER were calculated for these vowels. Subjects sung with equal or more power in the singer's formant region in choral versus solo mode in the context of the piece as a whole and in individual vowels. There was no difference in vibrato rate and extent between the two modes. Singing in choral mode, therefore, required the ability to use a similar vocal timbre to that required for solo opera singing.

Autophonic Loudness of Singers in Simulated Room Acoustic Environments.

OBJECTIVES: This paper aims to study the effect of room acoustics and phonemes on the perception of loudness of one's own voice (autophonic loudness) for a group of trained singers. METHODS: For a set of five phonemes, 20 singers vocalized over several autophonic loudness ratios, while maintaining pitch constancy over extreme voice levels, within five simulated rooms. RESULTS: There were statistically significant differences in the slope of the autophonic loudness function (logarithm of autophonic loudness as a function of voice sound pressure level) for the five phonemes, with slopes ranging from 1.3 (/a:/) to 2.0 (/z/). There was no significant variation in the autophonic loudness function slopes with variations in room acoustics. The autophonic room response, which represents a systematic decrease in voice levels with increasing levels of room reflections, was also studied, with some evidence found in support. Overall, the average slope of the autophonic room response for the three corner vowels (/a:/, /i:/, and /u:/) was -1.4 for medium autophonic loudness. CONCLUSIONS: The findings relating to the slope of the autophonic loudness function are in agreement with the findings of previous studies where the sensorimotor mechanisms in regulating voice were shown to be more important in the perception of autophonic loudness than hearing of room acoustics. However, the role of room acoustics, in terms of the autophonic room response, is shown to be more complicated, requiring further inquiry. Overall, it is shown that autophonic loudness grows at more than twice the rate of loudness growth for sounds created outside the human body.

Assessing Linearity in the Loudness Envelope of the Messa di Voce Singing Exercise Through Acoustic Signal Analysis.

Messa di voce (MDV) is a singing exercise that involves sustaining a single pitch with a linear change in loudness from silence to maximum intensity (the crescendo part) and back to silence again (the decrescendo part), with time symmetry between the two parts. Previous studies have used the sound pressure level (SPL, in decibels) of a singer's voice to measure loudness, so as to assess the linearity of each part-an approach that has limitations due to loudness and SPL not being linearly related. This article studies the loudness envelope shapes of MDVs, comparing the SPL approach with approaches that are more closely related to human loudness perception. The MDVs were performed by a cohort of tertiary singing students, recorded six times (once per semester) over a period of 3 years. The loudness envelopes were derived for a typical audience listening position, and for listening to one's own singing, using three models: SPL, Stevens' power law-based model, and a computational loudness model. The effects on the envelope shape due to room acoustics (an important effect) and vibrato (minimal effect) were also considered. The results showed that the SPL model yielded a lower proportion of linear crescendi and decrescendi, compared with other models. The Stevens' power law-based model provided results similar to the more complicated computational loudness model. Longitudinally, there was no consistent trend in the shape of the MDV loudness envelope for the cohort although there were some individual singers who exhibited improvements in linearity.

Towards a method for loudness-based analysis of the sound of one's own voice.

This paper outlines the steps in objectively estimating the time-varying loudness of one's own voice in a room (i.e. autophonic loudness). Voice recordings, made with a near-mouth microphone, are converted to the sound that reaches the two eardrums of the talking (or singing)-listener by convolving them with the impulse responses from the mouth to the respective ears of an anthropomorphic head and torso. The influences of bone-conducted sound and room reflections are taken into account. These convolved recordings are then processed with a computational time-varying loudness model. The method is demonstrated by a short case study, and the results illustrate something of the benefit of loudness analysis over sound pressure level analysis for representing autophonic loudness.

Change in messa di voce characteristics during 3 years of classical singing training at the tertiary level.

A 3-year longitudinal study was conducted to investigate changes in vocal quality as a result of singing training at a tertiary level conservatorium in Australia. Singers performed a messa di voce (MDV) at intervals of 6 months over the 3-year period of training. The study investigated the evolving acoustic features of the singers' voices exhibited during the MDV, including sound pressure level (SPL), short-term energy ratio (STER), duration, and vibrato parameters of the fundamental frequency (F0), SPL, and STER. The maximum SPL exhibited a marginal systematic increase over the training period, but the maximum STER did not systematically change. F0 vibrato extent increased significantly, whereas the extent of SPL and STER vibrato did not change significantly.

Long-term horizontal vocal directivity of opera singers: effects of singing projection and acoustic environment.

Vocal directivity refers to how directional the sound is that comes from a singer's mouth, that is, whether the sound is focused into a narrow stream of sound projecting in front of the singers or whether it is spread out all around the singer. This study investigates the long-term vocal directivity and acoustic power of professional opera singers and how these vary among subjects, among singing projections, and among vastly different acoustic environments. The vocal sound of eight professional opera singers (six females and two males) was measured in anechoic and reverberant rooms and in a recital hall. Subjects sang in four different ways: (1) paying great attention to intonation; (2) singing as in performance, with all the emotional connection intended by the composer; (3) imagining a large auditorium; and (4) imagining a small theatre. The same song was sung by all singers in all conditions. A head and torso simulator (HATS), radiating sound from its mouth, was used for comparison in all situations. Results show that individual singers have quite consistent long-term average directivity, even across conditions. Directivity varies substantially among singers. Singers are more directional than the standard HATS (which is a physical model of a talking person). The singer's formant region of the spectrum exhibits greater directivity than the lower-frequency range, and results indicate that singers control directivity (at least, incidentally) for different singing conditions as they adjust the spectral emphasis of their voices through their formants.

Effects of training on time-varying spectral energy and sound pressure level in nine male classical singers.

The male classical singing voice is a musical instrument that is very important in western culture. It has many acoustic features which should change and improve over the period in which the singer trains. In this study we compare nine singers in different stages of training, from university level students through to international soloists. Typically, Energy Ratio (ER; a measure of mean spectral slope) and mean sound pressure level (SPL) may be calculated to summarize an entire singing sample. We investigate an alternative approach, by calculating the time-varying ER and SPL. The inspection of the distribution of these descriptors over an aria's time period yields a more detailed picture of the strategies for high-frequency energy production used by singers with different levels of training.

Self-to-other ratios measured in an opera chorus in performance.

Four volunteer members of the chorus of Opera Australia, representing four different voice categories, wore binaural pairs of wireless microphones during a penultimate dress rehearsal on the Opera Theater stage of the Sydney Opera House. From the recordings, data were obtained on sound levels and on the self-to-other ratios (SORs). The sound levels were comparable to those found in loud music in chamber choir performance. The average SOR ranged from +10 to +15 dB. Compared to chamber choirs in other types of room, the SOR values were high. On a separate occasion, the stage support parameters ST1 (early reflections) and ST2 (late reflections) were measured over the whole stage area. ST1 was about -16 dB, which is typical for opera stages, and -20 dB for ST2, which is unusually low. It is concluded that the SOR in the opera chorus depends mostly on choir formation, which is highly variable, and that an opera chorus artist generally can hear his or her own voice very well, but little of the others and of the orchestra. This was confirmed by informal listening to the recordings.