Showing papers in "Journal of the Acoustical Society of America in 1972"

Speed of Sound in Pure Water

Abstract: A sound‐speed equation of fifth order in temperature is fit with a standard deviation of 0.0028 m/sec to 148 observations between 0.001°C and 95.126°C on the T68 scale. The accuracy is believed to be 0.015 m/sec, and the reproducibility over replications is 0.005 m/sec.

Emotions and Speech: Some Acoustical Correlates

Abstract: This paper describes some further attempts to identify and measure those parameters in the speech signal that reflect the emotional state of a speaker. High‐quality recordings were obtained of professional “method” actors reading the dialogue of a short scenario specifically written to contain various emotional situations. Excerpted portions of the recordings were subjected to both quantitative and qualitative analyses. A comparison was also made of recordings from a real‐life situation, in which the emotions of a speaker were clearly defined, with recordings from an actor who simulated the same situation. Anger, fear, and sorrow situations tended to produce characteristic differences in contour of fundamental frequency, average speech spectrum, temporal characteristics, precision of articulation, and waveform regularity of successive glottal pulses. Attributes for a given emotional situation were not always consistent from one speaker to another.

Attenuation of Sound in Suspensions and Emulsions: Theory and Experiments

Abstract: The absorption of sound in emulsions and suspensions is due to viscous and thermal transport processes occurring at the interface of the nonhomogeneities, as well as to the intrinsic absorption in the materials comprising the system A complete description of these processes for systems of fluid particles suspended in fluid media was given by Epstein and Carhart [J Acoust Soc Amer 25, 553 (1953)] However, subsequent investigators of liquid systems have generally neglected the thermal transport process We show in this work that the Epstein‐Carhart results provide a good description of our experimental attenuation data in emulsions, and that the thermal transport process can be the major factor in the attenuation We have here also extended the theory further to include the attenuation in suspensions of solid particles, and good agreement is found between our theory and experimental results for aqueous suspensions of polystyrene spheres

Radiation Resistance of a Rectangular Panel

Abstract: The radiation resistance corresponding to the natural modes of a finite rectangular panel is theoretically determined from the total energy radiated to the farfield. The panel is assumed to be simply supported in an infinite baffle. Asymptotic solutions for the low‐frequency region are derived, and curves covering the entire frequency range for various mode shapes and aspect ratios are obtained through numerical integration. When the ratio of the acoustic wavenumber to the panel wavenumber is a constant much less than unity, the radiation resistance for all modes is a minimum if the intranodal area (the area between adjacent node lines) is square, and increases with the aspect ratio of the intranodal area.

Perceived Level of Noise by Mark VII and Decibels (E)

Abstract: The calculation procedure Mark VII gives the perceived level of loudness or noisiness in PLdB. It utilizes a set of frequency‐weighting contours based on an average of 25 experimental contours. The standard reference sound is defined as a 13‐oct band centered at 3150 Hz. The perceived magnitude (loudness or noisiness) grows as the 23 power of the sound pressure, so that perceived magnitude doubles with each increase of 9 dB. The summation formula for the total subjective magnitude remains St = Sm + F (∑S − Sm), but the value of F is made to vary as a function of level in order to reflect the nonlinear growth (in log‐log coordinates) of broad‐band noise. As a result of the new reference sound at 3150 Hz, perceived level in decibels (PLdB) is approximately 8 dB lower than the older loudness level in phons. Except for the nearly constant difference of 8 dB, Mark VI and Mark VII give closely similar results for typical broad‐band noises. The 8‐dB downward shift makes it possible for a sound level meter with a...

The Timing of Utterances and Linguistic Boundaries

Abstract: This paper is concerned with the effect of morphological and syntactic boundaries on the temporal structure of spoken utterances. Two speakers produced 20 tokens each of four sets of words consisting of a mono‐syllabic base form, disyllabic and trisyllabic words derived from the base by the addition of suffixes, and three short sentences in which the base form was followed by a syntactic boundary; this in turn was followed by a stressed syllable, one unstressed syllable, and two unstressed syllables. The sentences thus reproduced the syllabic sequences of the derived words. The duration of words and segments was measured from oscillograms. The manifestation of morphological and syntactic boundaries is discussed, and some implications of the findings relative to the temporal programming of spoken utterances are considered.

Preceding Vowel Duration as a Cue to the Perception of the Voicing Characteristic of Word‐Final Consonants in American English

Abstract: A number of studies in the literature have stated that the duration of a vowel is a significant cue to the voicing characteristic of the consonant that follows it. The present study investigated the effect of varying preceding vowel duration upon the perception of word‐final stops, fricatives, and clusters in synthetic speech. A variety of minimal CVC (C) pairs was synthesized and the vowel of each was varied over a range of values derived from durations found in real‐speech samples. It was found that, regardless of the cues for voicing or voicelessness used in the synthesis of the final consonant or cluster, listeners perceived the final segments as voiceless when they were preceded by vowels of short duration and as voiced when they were preceded by vowels of long duration. Discrimination tests revealed that when the voicing characteristic is cued by vowel duration, perception is continuous rather than categorical.

Intensity Perception. II. Resolution in One‐Interval Paradigms

Abstract: This paper reports the results of a series of experiments on tone pulses designed to test certain predictions of the preliminary theory of intensity resolution (Durlach and Braida, 1969) relevant to one‐interval paradigms. Resolution was measured in identification and scaling experiments as a function of the range, number, and distribution of intensities, and the availability of feedback. Some of the results, such as those on the dependence of resolution on range and number of stimuli in absolute identification, support the theory. Other results, however, such as those comparing resolution in identification with resolution in magnitude estimation for a small common range, indicate that the theory is inadequate and needs to be revised.

The Central Origin of the Pitch of Complex Tones: Evidence from Musical Interval Recognition

Abstract: The human auditory system's ability to recognize simple melodies that correspond to fundamental periods in sequences of periodic sounds devoid of fundamental energy was studied through musical interval identification experiments. Stimuli comprising two randomly chosen successive upper harmonics were presented both monotically (two harmonics to one ear) and dichotically (one harmonic to each ear). Subjects could recognize melodies equally well with both modes of stimulus presentation. The results imply that the pitch of these complex tones is mediated by a central processor operating on neural signals derived from those effective stimulus harmonics that are tonotopically resolved.

Psychophysical evidence for lateral inhibition in hearing.

Abstract: Although there are some indications of the possible role of lateral inhibition in hearing, there has been no clear demonstration of it in psychophysical experiments. Either the phenomenon plays only a minor role, or it has escaped psychophysical verification. Accepting for a moment the second possibility, is argued in this paper that the threshold of a test tone presented simultaneously with a masker does not reflect clear lateral‐inhibition effects since the inhibition affects both the test tone and the masker. Two different methods, in which the test tone and the masker were presented succsesively, give clear psychophysical evidence of lateral inhibition in hearing. Firstly, the threshold curve of short test‐tone bursts presented in the gaps between repeated masker bursts (noise with a steep negative or positive gradient at a particular frequency) shows marked edge effects. Secondly, the results of psychophysical measurements on two‐tone suppression indicate that the nervous activity due to one frequency component may be suppressed by another component. The effect at the edges of the frequency spectrum are comparable with visual Mach bands, and the interaction of two tones is suggestive of the two‐tone inhibition found in auditory‐nerve fibers.

An Experimental and Theoretical Investigation of Elastic Wave Propagation in a Cylinder

Abstract: A high‐speed computer was used to investigate the problem of wave propagation in an isotropic elastic cylinder. Dispersion curves corresponding to real, imaginary, and complex propagation constants for the symmetric and the first four antisymmetric modes of propagation are given. The radial distributions of axial and radial displacements and of shear and normal stresses are given for the symmetric mode. By using a finite number of modes of propagation, an approximate solution is found for the problem of the L(0,1) mode impinging on a traction‐free interface. The reflection coefficient is determined in this way and the accompanying generation of higher order modes at the interface is shown to cause a high‐amplitude end resonance. Experimental results obtained by using the resonance method in conjunction with a long rod are presented to substantiate the calculated reflection coefficient and the frequency of end resonance. Phase velocities, based on measurements of the wavelength of standing waves and resonance frequencies, were obtained for the symmetric and first two antisymmetric modes. These measurements extend into the frequency range of more than one propagating mode. The rms deviation between theoretical and experimental results is in general less than 0.2% with the exception of the dispersion curve for the L(0,2) mode which deviates by 0.7%.

Noise reduction by vegetation and ground

Abstract: Transmission of random noise through dense corn, a dense hemlock plantation, an open pine stand, dense hardwood brush, and over cultivated soil was measured. The relation between attenuation and frequency in these diverse cases suggested models that permit the prediction of attenuation in any configuration of vegetation and soil. The corn crop had an excess attenuation of 6 dB/100 ft for each doubling of frequency between 500 and 4000 Hz. On the other hand, the stems of the hemlock, pine, and brush all reduced noise by only about 5 dB/100 ft at 4000 Hz. Bare ground attenuates frequencies of 200–1000 Hz, and the frequency of maximum attenuation depends on the soil permeability to air. Thus, tilling the soil reduced the frequency of peak attenuation from 700 to 350 Hz and increased maximum attenuation at 52 m from the source by nearly 80%. Furthermore, earlier conflicting reports of noise attenuation by vegetation appear reconciled if ground attenuation is taken into account. Scattering and ground attenuation are the principal factors in sound attenuation by vegetation. Both factors attenuate relatively less sound as distance from the sound source increases. Hence measurements far from the source can underestimate the effect of a narrow band of vegetation or soil.

Tympanic-membrane vibrations in human cadaver ears studied by time-averaged holography.

Abstract: Tympanic‐membrane (TM) vibrations were studied in fresh human‐cadaver specimens by means of time‐averaged holography. As with an earlier study in cats [S. M. Khanna and J. Tonndorf, J. Acoust. Soc. Amer. 51, 1904 (1972)], the findings supported the curved‐membrane hypothesis of Helmholtz. Despite a careful search for experimental artifacts, no satisfactory explanation was found to account for the fact that the present results are at variance with the stiff‐plate concept of von Bekesy. With respect to their quantitative evaluation, the human‐cadaver data were not as clear‐cut as the earlier cat data on account of anatomical differences. A number of points, which on first glance do not seem to agree with Helmholtz's concept are being discussed in the light of present or other recent findings, and explanations are being offered. These topics include: (1) the compliance of the TM; (2) the effects of differential air pressure across the TM upon middle‐ear transmission; (3) the effects of TM perforations; and (...

Efficient Acoustic Parameters for Speaker Recognition

Abstract: In a scheme for the mechanical recognition of speakers, it is desirable to use acoustic parameters that are closely related to voice characteristics that distinguish speakers. This paper describes an investigation of an efficient approach to selecting such parameters, which are movitated by known relations between the voice signal and vocal‐tract shapes and gestures. Rather than general measurements over the extent of an utterance, only significant features of selected segments are used. A simulation of a speaker recognition system was performed by manually locating speech events within utterances and using parameters measured at these locations to classify the speakers. Useful parameters were found in fundamental frequency, features of vowel and nasal consonant spectra, estimation of glottal source spectrum slope, word duration, and voice onset time. These parameters were tested in speaker recognition paradigms using simple linear classification procedures. When only 17 such parameters were used, no errors were made in speaker identification from a set of 21 adult male speakers. Under the same conditions, speaker verification errors of the order of 2% were also obtained.

Tympanic Membrane Vibrations in Cats Studied by Time‐Averaged Holography

Abstract: The vibratory characteristics of the cat's tympanic membrane (TM) were studied in fresh cadavers and living animals with the aid of time‐averaged holography. Initially, technical difficulties were presented by the TM's small size, low stiffness, and its transparency. Time‐averaged holograms of the TM were obtained for various sound‐pressure levels at a number of frequencies between 600 and 6000 Hz. Reconstructions of such holograms show an image of the TM superimposed by isoamplitude contour lines giving the absolute displacement amplitude at any point on the TM. Vibratory patterns remain essentially unchanged in their first mode up to a frequency of 2000 Hz, with higher modes occurring above that value. At all frequencies, the TM vibrates maximally in the posterior‐superior quadrant and less in the anterior and inferior ones. These results show that, even in its first mode of operation, i.e., below 2000 Hz, the TM does not vibrate like a stiff plate as observed by von Bekesy in human cadavers. Rather, the present findings support the curved‐membrane concept of Helmholtz. Contributions to the total transformer action of the middle ear by the area ratio: TM/stapes footplate, the ossicular lever ratio, and that due to the curved membrane action are reevaluated.

Duration discrimination of noise and tone bursts.

Abstract: This research investigates the human observer's ability to discriminate difference in duration for noise bursts and gated sinusoids. Two observers compared two durations T and T+ΔT in a two‐alternative forced‐choice procedure. The value of T ranged from 0.16 to 960 msec. For each T the value of ΔT for 75% discrimination was determined. The results showed that for most of the range investigated ΔT was proportional to T12. Performance was not affected by a change in bandwidth from 3500 to 200 Hz. Values of ΔT for 75% correct did decrease, however, when the observers were given audible spectral cues from very short pulsed sinusoids. The theory best describing the results was the neural counter model proposed by Creelman [J. Acoust. Soc. Amer. 34, 582–593 (1962)].

Combination Tones and Their Origin

Abstract: Combination tones corresponding to f1 − k (f2 − f1), with k a small positive integer, are often audible during stimulation by the two frequency components f1 and f2 (f1 < f2). In this paper primarily the cubic difference tone (CDT) 2f1 − f2 (k = 1) is studied. The research reported is directed at the type of nonlinearity generating the CDT, and at the site of CDT generation. The first experiment shows that the generation of the CDT is affected by a dip (a threshold elevation in a narrow frequency region). The CDT was perceived only when the level of f2 exceeded the elevated threshold. In the second experiment the cancellation method is used. The results suggest that the high‐frequency slope of the pattern of stimulation upon which the nonlinearity operates is comparable with the slope revealed in masking experiments. In the last experiments the cancellation method is reconsidered. Estimates of the CDT level found by various other measuring methods, in which the probe tone was presented nonsimultaneously w...

Is the left hemisphere specialized for speech, language and-or something else?

Abstract: Experienced Morse code operators showed significant right‐ear superiority, indicating left‐hemisphere lateralization, for the perception of dichotically presented Morse code letters. No significant lateralization was found for rapid monotically presented Morse code words. Subjects who did not know Morse code did not show significant lateralization when dichotically presented with a set of stimuli which included all Morse code letters; however, they showed a tendency toward left‐hemisphere lateralization with a set of Morse code letters which were restricted in duration and presented at relatively low intensity. Our results show that articulability is not a necessary property of stimuli lateralized to the left hemisphere in dichotic listening; a possible further interpretation is that it is language rather than speech that is lateralized to the left hemisphere.

Sound Radiated by Turbulent Flow off a Rigid Half-Plane as Obtained from a Wavevector Spectrum of Hydrodynamic Pressure

Abstract: The spectrum and intensity are calculated for radiated pressure generated from turbulent flow by diffraction at the edge of a rigid half‐plane. The formulation relates the radiated spectrum to the local wavevector‐frequency spectral density of pressure near the edge in a plane bounding the flow and is based on the notion of a hydrodynamic pressure field driving a contiguous sound field. In the half‐plane extension, the pressure diffracted by the rigid surface vanishes, and the spectrum of the driving hydrodynamic pressure may be approximated accordingly. Results are obtained without explicit reference to fluctuating velocity when the (low Mach number) flow is regarded as confined to one side of the half plane and a model suitable for turbulent‐boundary‐layer wall pressure is used for the driving spectrum. Results are also obtained from a model suitable for pressure in the inertial subrange of homogeneous turbulence. From the wall‐pressure model, the frequency spectrum of radiated pressure, for ωδ/U⩾5 cosα...

An Analysis of Lesion Development in the Brain and in Plastics by High‐Intensity Focused Ultrasound at Low‐Megahertz Frequencies

Abstract: Thermal factors are believed to play a dominant role in the development of the structural and functional effects of irradiation of the nervous system with focused ultrasound at low‐megahertz frequencies. Similar mechanisms are postulated to underlie the effects of irradiation in methacrylate, which is frequently used as a test material to evaluate the influence of various factors on the results obtained. This study was undertaken to determine if thermal mechanisms alone can explain the development of trackless focal alterations (lesions) and all of their measurable characteristics in plastic as well as in brain. A purely thermal model is assumed and analytical prediction of lesion development and lesion size and shape for varying values of ultrasonic and thermal constants and controllable variables (frequency, focusing, dosage, target depth, etc.) is attempted. An empirical equation to describe the axial and radial ultrasonic energy distribution at the focus in water is derived. Appropriate heat transfer ...

Reflection‐Refraction of a Stress Wave at a Plane Boundary between Anisotropic Media

Abstract: A review is given of the pertinent equations necessary to describe the reflected and refracted waves at a plane boundary between anisotropic media and the utility of the wave surface in discussing this problem. The critical angle phenomenon in anisotropic media is discussed in terms of the energy flux vector associated with the reflected and refracted modes. The critical angle is shown to occur generally at that angle of incidence for which the energy flux vector of the reflected or refracted mode is parallel to the boundary and not when the wave vector is parallel to the boundary. The possibility of not needing a nonhomogeneous surface wave to satisfy the boundary conditions at angles of incidence greater than the critical angle is discussed for certain particular regions in some anisotropic materials.

Identification of Stops and Vowels for the Burst Portion of /p, t, k/ Isolated from Conversational Speech

Abstract: Stimulus segments (burst, and burst plus 100 msec of vowel) were isolated from words with initial and final /p,t,k/, and presented to listeners for identification. For the burst alone, initials were identified better than finals and /t/ showed the highest level of identification. Also, in many instances vowels could be identified on the burst portions, giving perceptual evidence for coarticulation. With the additional 100 msec, the level of identification, as was expected, was high. Providing phonotactic context, in some instances, raised this level of identification.

Transmission of Sound in Ducts with Thin Shear Layers—Convergence to the Uniform Flow Case

Abstract: The problem of the transmission of sound in a duct with very thin shear layers at the walls is treated by an inner expansion method. The results show that the formulation of the problem of the transmission of sound in a duct with a shear layer at the wall converges, in the case of a vanishingly thin shear layer, to the formulation of the same problem when uniform flow is assumed and the wall boundary condition is that of continuity of particle displacement.

Atmospheric Absorption of Sound: Theoretical Predictions

Abstract: By assuming that air is composed of four gases (i.e., nitrogen, oxygen, water vapor, and carbon dioxide) and applying energy transfer rates for the binary collisions inherent in such a system, absorption of sound in the atmosphere has been predicted. The calculated curves based upon 24 energy transfer mechanisms are compared with experimental data over the humidity range of 0–100% relative humidity. Agreement between theory and experiment is very good. By including classical absorption and rotational relaxation effects, the total atmospheric absorption is also predicted at 20°C. Calculations made for various concentrations of CO2 indicate that low CO2 levels (less than 0.1%) do not significantly affect absorption of audible sound at high humidities. At very low humidities, however, CO2 is an important factor.

Direct Correlation of Noise and Flow of a Jet

Abstract: A direct correlation between the flow and noise of a jet—between the “cause” and the “effect”—has been measured. Two kinds of correlations were explored, namely (1) the broad‐band turbulence signal (hot‐film) with the broad‐band acoustic signal (microphone), and (2) the narrow‐band filtered turbulence signal with the narrow‐band filtered acoustic signal; the latter approach was ultimately adopted. The correlations were analyzed in terms of an extension of Proudman's form of Lighthill's integral for aerodynamic noise; this yielded the relative intensity and spectrum of the noise originating from unit volume of a jet (35 locations) and received at a farfield point (r = 96D, φ = 40°); this in turn led to the relative emission of successive “slices” of a jet versus axial distance X over the measurement range (1D ⩽ × ⩽ 7D). Qualitative agreement was found with Ribner's X° law, and the spectral peaks for each slice were located in frequency essentially as predicted by Powell. The maximum normalized broad‐band c...

Audibility region of combination tones

Abstract: This paper describes research on combination tones of the type f1 − k (f2 − f1. Such tones are audible during stimulation by the two frequency components f1 and f2, with f1 < f2. The results of the experiments suggest that these combination tones are audible only in a restricted frequency region below f1. The lower limit of this “audibility region,” for the combination tone 2f1 − f2 (k = 1), was determined primarily by the level of the lower stimulus component L1 and additionally by L2. The width of the audibility region appeared to be highly subject dependent. The lowest stimulus level for which 2f1 − f2 was audible was 15–20 dB SL if both stimulus components had equal levels. If the lower component was set at 40 dB, then the higher one could be reduced to 4 dB before 2f1 − f2 reached threshold. Higher‐order combination tones could be perceived up to k = 5 or 6. At 40 dB SL, the lower limit of the audibility region was approximately the same regardless of the order (i.e., independent of k). The width of the audibility region could not be explained completely by the amount of interaction of the mechanical stimulation patterns corresponding to f1 and f2. In addition, it is demonstrated that these combination tones may play a misleading part in masking experiments. The detection threshold of a simple tone masked by another simple tone may be determined by the detection of a combination tone of the type f1 − k (f2 − f1), even at a low level of the masker.

Shear as a mechanism for sonically induced biological effects.

Abstract: By controlled irradiation of erythrocyte suspensions at 20 kHz it is demonstrated that shear associated with acoustic microstreaming can be an important mechanism for biological effects of sound. Two effective sources of acoustic microstreaming are stable oscillating gas bubbles and transversely oscillating wires. The threshold displacement amplitude for achieving critical shear can be reduced by increasing the solvent viscosity and reducing the radius of the source of acoustic streaming. The threshold stress was found to decrease by 55% or more when the sample was heated to 45°C or higher for 10 min. This suggests that synergism exists between mechanical and thermal mechanisms for sonic effects. Mass transfer associated with small‐scale acoustic streaming controls the rate of cell disruption.

Auditory Thresholds of a Killer Whale Orcinus orca Linnaeus

Abstract: Using standard operant conditioning techniques, a killer whale, Orcinus orca Linnaeus, was trained to respond to pure‐tone auditory signals by pushing a response manipulandum. An audiogram was obtained for frequencies between 500 Hz and 31 kHz. Greatest sensitivity to the signal was observed at 15 kHz at a level of −70±5 dB re 1 dyn/cm2. The observed upper limit of hearing was 32 kHz. At no time during training or testing did the animal respond to an undistorted signal above 32 kHz. Frequencies below 500 Hz were not tested, owing to high ambient tank noise levels.

Intensity Discrimination of Pulsed Sinusoids: The Effects of Filtered Noise

Abstract: It has been shown repeatedly that Weber's law does not hold for intensity discrimination of pulsed tones. The masking function for such waveforms (10 logΔI vs 10 logI, for constant performance) typically is linear with a slope of 0.9 over a range in I of at least 60 dB. The present experiment indicates that the slope of the masking function increases from 0.9 to 1.0 as regions above the signal frequency are masked by filtered noise. A sufficient condition for a slope of 1.0 is a high‐pass noise which masks regions at and above the second harmonic of the signal. This suggests that the “near miss” to Weber's law reflects the observer's use of information at aural harmonics. A model based upon this notion gives a good quantitative account of the data. The amount of distortion assumed in the model is in reasonable agreement with other data.

Duration as a Cue in the Recognition of Synthetic Vowels

Abstract: Sets of synthetic vowel sounds which differed in first‐formant frequency, second‐formant frequency, and duration were presented to a group of listeners for identification. It was found that listeners' judgments depended on all of these factors. Duration was a relatively more important cue for vowels located in the center of the F1‐F2 space where a vowel might more readily be confused with one of its neighbors. The perceived duration of a vowel was found to be biased by the rhythm of the sounds that preceded it. In the case of sounds lying near perceptual boundaries, this was sometimes sufficient to change the identity of the vowel.