1、 Access to Additional Content for ASA S3.4:2007 (Click here to view the publication) This Page is not part of the original publication This page has been added by IHS as a convenience to the user in order to provide access to additional content as authorized by the Copyright holder of this document
2、Click the link(s) below to access the content and use normal procedures for downloading or opening the files. ANSI S3.4-2007 Loudness program Information contained in the above is the property of the Copyright holder and all Notice of Disclaimer Glasberg and Moore, 2006) that was developed from a mo
3、del originally proposed by Zwicker and his co-workers (1958, 1965, 1984, 1999). Zwickers model is part of ISO 532:1975. These models have their roots in the pioneering work of Fletcher and Munson (1933). The current procedure extends and improves the accuracy of these earlier methods. It replaces th
4、e more limited computational procedure used in the old ANSI S3.4-1980 that was based on the method proposed by S.S. Stevens (1957, 1961). Unlike the old ANSI S3.4-1980, the current standard can be applied to sounds with sharp line spectral components, e.g., transformer hum or fan noise, as well as t
5、o sounds with broadband spectra. Provided the measurement of sound pressure levels is sufficiently precise, and subject to certain restrictions specified below, the recommended procedure may also be used to estimate loudness, or loudness level, with reasonable accuracy down to near threshold levels.
6、 Moreover, it enables the loudness of complex sounds containing spectral energy below 500 Hz to be determined. The equal-loudness contours derived from this standard are in good agreement with ISO 226:2003 (see also Suzuki and Takeshima, 2004). Because loudness is a subjective quantity, the percepti
7、on of which may vary among people, any calculated loudness value represents only an estimate of the average loudness as perceived by a group of individuals with normal hearing. The changes made in this revision are: (1) A modification to the method for calculating specific loudness from excitation f
8、or center frequencies below 500 Hz to make the standard fully compatible with the provided software and with the model of Moore et al. (1997) and with its amendment by Glasberg and Moore (2006); (2) A modification to the transfer function of the middle ear, as proposed by Glasberg and Moore (2006),
9、which allows more accurate predictions of the absolute thresholds in ISO 389-7:2005 and of the equal-loudness contours in ISO 226:2003. The software provided with this American National Standard is entirely informative and provided for the convenience of the user. Use of the provided software is not
10、 required for conformance with the Standard. ASA and the owners of the copyright to the software provided with this American National Standard make no other representation or warranty or condition of any kind, whether express or implied (either in fact or by operation of law) with respect to any par
11、t of the product, including, without limitation, with respect to the sufficiency, accuracy or utilization of, or any information or opinion contained or reflected in, any of the product. ASA and the owner expressly disclaim all warranties or conditions of merchantability or fitness for a particular
12、purpose. No officer, director, employee, member, agent, 2007 Acoustical Society of America. All rights reserved. iv representative or publisher of the copyright holder is authorized to make any modification, extension, or addition to this limited warranty. At the time this Standard was submitted to
13、Accredited Standards Committee S3, Bioacoustics for approval, the membership was as follows: C.A. Champlin, Chair R.F. Burkard, Vice-Chair S.B. Blaeser, Secretary Acoustical Society of America . C.A. Champlin R.F. Burkard (Alt.) American Academy of Audiology .Y. Szymko-Bennett D.A. Fabry (Alt.) Amer
14、ican Academy of Otolaryngology, Head and Neck Surgery, Inc. . .R.A. Dobie . L.A. Michael (Alt.) American Industrial Hygiene Association T.K. Madison . D. Driscoll (Alt.) American-Speech-Language-Hearing Association (ASHA) L.A. Wilber V. Gladstone (Alt.) Beltone/GN Resound S. Petrovic Council for Acc
15、reditation in Occupational Hearing Conservation (CAOHC) .E.H. Berger .J.A. Mann (Alt.) Etymotic Research, Inc. M.C. Killion .R. Scicluna (Alt.) Food and Drug Administration J. Kane Frye Electronics, Inc. G.J. Frye K.E. Frye (Alt.) Hearing Industries Association .T.A. Victorian .C.M. Rogin (Alt.) Nat
16、ional Hearing Conservation Association T. Schulz National Institute for Occupational Safety and Health M. Stephenson .W.J. Murphy (Alt.) National Institute of Standards and Technology V. Nedzelnitsky R. Wagner (Alt.) Quest Technologies, Inc.M. Wurm .P. Battenberg (Alt.) Starkey Laboratories D.A. Pre
17、ves . T. Burns (Alt.) 2007 Acoustical Society of America. All rights reserved. vU.S. Air Force . R. McKinley U.S. Army Aeromedical Research Lab W. Ahroon U.S. Army Construction Engineering Research LaboratoriesL. Pater .D. Delaney (Alt.) U.S. Army Human Research FAX: 631-390-0217; E-mail: asastdsaip
18、.org AMERICAN NATIONAL STANDARD ANSI S3.4-2007 2007 Acoustical Society of America. All rights reserved. 1American National Standard Procedure for the Computation of Loudness of Steady Sounds 1 Scope This standard specifies a procedure for calculating the monaural and binaural loudness experienced by
19、 listeners with normal hearing under the following conditions: 1.1. Listening conditions. The standard applies to three listening conditions: free field with a frontal incidence, diffuse field, or listening via headphones. Listening in a free field with a frontal incidence is assumed when the sound
20、source is located at least one meter directly in front of the listener and there are no space boundaries or other surfaces that affect the sound field. The free-field condition can be achieved in an anechoic chamber or in open space. Listening in a diffuse field is assumed when the sound reaches the
21、 listeners ears from all directions with essentially the same acoustic power. The sound field in a room with hard reflective walls where the sound is reflected many times before being significantly absorbed approximates a diffuse field. Listening through headphones is assumed when the sound is deliv
22、ered directly to the listeners ears through circumaural or supra-aural headphones, or insert earphones. For simplicity, the term headphones will be used hereafter to denote both headphones and earphones. NOTE 1 The equal-loudness contour for narrow-band noise in a free field differs from that in the
23、 diffuse field, but the difference measured at 1/3-octave-band center frequencies varies from -2.0 to +4.3 decibels with the mean difference equal to only 0.8 decibels. These differences do not significantly affect calculations for broadband spectra. Nevertheless, since the loudness of a sound depen
24、ds on the nature of the enclosure in which it is heard, comparative evaluations of different sources should be based on measurements all made in essentially similar enclosures or all made in a free sound field. NOTE 2 Pressure levels in a diffuse field should be measured by means of an omnidirection
25、al microphone located in the unobstructed sound field at the position corresponding to the center of the listeners head. 1.2 Spectrum. The computational procedure described in the standard applies to a variety of sounds including complex tones, noise bands, and mixtures of the two. The characteristi
26、cs of the sounds are specified in terms of their spectra. The procedure applies to sounds with a frequency range that extends from 20 Hz to 18,000 Hz. However, the procedure may not give accurate estimates of loudness for sounds with strong components above 12,500 Hz, and perceived loudness for such
27、 sounds is likely to vary markedly across individuals. 1.3 Steady state. The procedure described in the standard applies to steady state sounds and should not be used for impulse sounds or intermittent sounds. Application to such sounds may lead to discrepancies between measured and calculated loudn
28、ess levels. The magnitude of the discrepancy is related to the dynamic characteristics of the device used to determine the sound pressure levels. 2 Terms and definitions For the purposes of this standard, the terms and definitions given in ANSI S1.1-1994 and ANSI S3.20-1995 and the following apply:
29、ANSI S3.4-2007 2007 Acoustical Society of America. All rights reserved. 2 2.1 sound pressure level. Ten times the logarithm to the base ten of the ratio of the time-mean-square pressure of a sound, in a stated frequency band, to the square of the reference sound pressure in gases of 20 Pa. Unit, dec
30、ibel (dB); abbreviation, SPL; symbol, Lp. 2.2 band pressure level. Sound pressure level for sound contained within a restricted frequency band. Unit, decibel (dB); abbreviation, BSPL; symbol, Lpb. NOTE A band may be identified by its nominal lower and upper band edge frequencies, or by its nominal m
31、idband frequency and nominal bandwidth. The nominal width of the band may be indicated by a prefactory modifier, e.g., octave-band sound pressure level, one-third-octave-band sound pressure level, or 50-Hz band level at 400 Hz. 2.3 spectrum density level; spectrum level. Level of the limit, as the w
32、idth of the frequency band approaches zero, of the quotient of a specified quantity distributed within a frequency band, by the width of the band. The words “spectrum level“ shall be preceded by a descriptive modifier. Unit, decibel. NOTE For illustration, sound pressure spectrum level Lpsat the mid
33、band frequency is obtained practically by ( ) ( ) fpfpLps 020210/log10 = , where p2is the time-mean-square sound pressure measured through a filter system, p0the reference sound pressure, f the nominal bandwidth of the filter system, 0f the reference bandwidth of 1 Hz. For computational purposes, wi
34、th Lpbfor the band pressure level observed through the filter, the above relation becomes ffLLpbpbs 010/log10 = . 2.4 loudness level. The loudness level of a sound is equal to the SPL of a 1000-Hz sinusoid that is judged equally loud. The 1000-Hz sinusoid is presented in a free field with frontal in
35、cidence and is heard with both ears. The unit of loudness level is the phon. NOTE The manner of listening to the unknown sound should be specified. 2.5 loudness. A numerical designation of the strength, in sones, of a sound which is proportional to the subjective magnitude as estimated by listeners
36、having normal hearing. One sone is the loudness of a sound whose loudness level is 40 phons. NOTE Loudness depends primarily upon the sound pressure although it also depends upon the frequency, waveform, bandwidth, and duration of the sound. 2.6 excitation level. The calculated level at the output o
37、f an auditory filter centered at the frequency of interest. The level is specified in decibels relative to the level produced by a 1000-Hz tone at 0 dB SPL presented in a free field with frontal incidence. 2.7 specific loudness. The loudness evoked over a one-ERBN-wide range of center frequencies ce
38、ntered on the frequency of interest, where ERBNdenotes the average value of the equivalent rectangular bandwidth of the auditory filter as determined for young normally hearing listeners at a moderate sound level. Specific loudness is a loudness density with units of sones/ERBN. ANSI S3.4-2007 2007
39、Acoustical Society of America. All rights reserved. 33 Procedure 3.1 Introduction The calculation of loudness is based on a computational procedure similar, but not identical, to that described by Moore, Glasberg and Baer (1997) and Glasberg and Moore (2006). This procedure is more comprehensive and
40、 accurate than the procedure used in the old ANSI S3.4-1980. For a more detailed historical perspective, see the Foreword on Page iii. When the term “files” is used in the text, this refers to files that are used within the computational procedure. It is envisaged that those wishing to calculate lou
41、dness using the procedure will use a computer program that implements the procedure. One such program accompanies this standard. The procedure involves a sequence of stages. Each stage is described below. The description is given so that the reader can understand how the procedure works. However, it
42、 is not expected that the procedure will be implemented “by hand.” Such computations would be very time consuming. 3.2 Specification of input signal The spectrum of the signal whose loudness is to be determined shall be specified. This can be done in several ways as described below. (1) The spectrum
43、 can be specified in terms of the frequency components present in a complex tone. The components can be either harmonically or non-harmonically spaced. The frequency in Hz and SPL in dB of each component shall be specified. (2) The spectrum can be specified in terms of bands of noise of defined widt
44、h. The number of noise bands shall also be specified. Each band can be composed of either white noise (with a constant spectrum level within the passband) or pink noise (whose spectrum level within the passband decreases with increasing center frequency at a rate of 3 dB/octave). For each band, the
45、following shall be specified: the lower cutoff frequency, the upper cutoff frequency, and the spectrum level. In the case of pink noise, the frequency at which the spectrum level is determined shall also be specified. Within the procedure, the spectra of bands of noise are approximated by a series o
46、f discrete sinusoidal components. When the bandwidth of the noise exceeds 30 Hz, the components are spaced at 10-Hz intervals, and the level of each component is set 10 dB higher than the spectrum level at the corresponding frequency. For example, a band of white noise extending from 200 to 500 Hz w
47、ith a spectrum level of 50 dB would be approximated by sinusoidal components with frequencies 205, 215, 225, 235 . 475, 485, 495 Hz, each component having a level of 60 dB SPL. When the bandwidth of the noise is less than 30 Hz, the components are spaced at 1-Hz intervals, and the level of each comp
48、onent is set equal to the spectrum level at the corresponding frequency. NOTE The spacing of the components (10 Hz or 1 Hz) is not a property of the input signal. The spacing is chosen to approximate the spectrum of the signal with sufficient accuracy for the purpose of the computation of loudness.
49、(3) The spectrum can be specified in terms of a mixture of discrete sinusoidal components and bands of noise. Point (1) above specifies the former and (2) specifies the latter. (4) The spectrum can be specified in terms of the sound pressure levels in 26 adjacent 1/3 octave bands with nominal center frequencies of 50, 62.5, 80, 100, 125, 160, 200, 250, 315, 400, 500, 630, 800, 1000, 1250, 1600, 2000, 2500, 3150, 4000, 5000, 6300, 8000, 10000, 12700 and 16000 Hz (exact center frequencies are 49.6, 62.5, 78.7, 99.2, 125, 157.5, 198.4, 250, 315, 396.9, 500,
