• Volume/Page
  • Keyword
  • DOI
  • Citation
  • Advanced
   
 
 
 

Journal of the Acoustical Society of America

SECTION LISTING

BROWSE VOLUMES

Year Range: 
Search Issue | RSS Feeds RSS
Previous Issue

Jun 1970

Volume 47, Issue 6B, pp. 1509-1618


Intensity Changes at the Ear as a Function of the Azimuth of a Tone Source: A Comparative Study

J. M. Harrison and P. Downey

J. Acoust. Soc. Am. Volume 47, Issue 6B, pp. 1509-1518 (1970); (10 pages) | Cited 1 time

Online Publication Date: 11 Aug 2005

Full Text: | Download PDF

Show Abstract
Sound intensity of a tone (4, 10, and 20 kHz) was measured at the external auditory meatus as a function of the azimuth (at every 15°) of the sound source. The observations were made under free‐field conditions, and man, squirrel monkey, rat, and bat were used as subjects. At 4 kHz, rat and man showed consistent variation of intensity as a function of azimuth. Only irregular variations were obtained in monkey and bat. Interaural‐intensity differences (IIDS) at 4 kHz were estimated to be sufficient to support localization of a tone at a minimum angle of 2.1° (rat) and 1.125° (man) from the midline. At 10 kHz, rat, bat, and monkey showed consistent variations of intensity with azimuth of at least 15 dB. IIDs at 10 kHz were estimated to be sufficient to support the localization of a tone at a minimum angle of 1.125° (monkey), 2.0° (bat), and 1.5° (rat) from the midline. At 20 kHz, consistent intensity variations of at least 14 dB were obtained in rat, bat, and monkey. IIDs were estimated as sufficient to support localization at a minimum angle of 1.5° (monkey), 2.25° (bat), and 3.6° (rat).

Effects of Electrical Current Applied to Cochlear Partition on Discharges in Individual Auditory‐Nerve Fibers. I. Prolonged Direct‐Current Polarization

Teruzo Konishi, Donald C. Teas, and Joel S. Wernick

J. Acoust. Soc. Am. Volume 47, Issue 6B, pp. 1519-1526 (1970); (8 pages)

Online Publication Date: 11 Aug 2005

Full Text: | Download PDF

Show Abstract
This study is designed to reveal the effects of electrical polarization applied to the cochlear partition on discharges in individual auditory‐nerve fibers in guinea pigs. Direct current is introduced in the basal turn across the organ of Corti. Its effects on the impulse discharges of the primary auditory fibers are measured during systematic variation of electrical and acoustic parameters. Data, recorded on tape, are analyzed by computing latency and interval histograms. With external current flowing from scala vestibuli to scala tympani, the spontaneous activity and sound‐evoked responses increase in most of the auditory‐nerve fibers. Current in the opposite direction causes decrease of impulse discharges. These nerve impulses show adaptation during application of the current and an aftereffect following its offset. It appears that electrical polarization modifies the resting current through the hair cells and also the excitability of the initial segment of the afferent auditory fibers.

Effects of Electrical Current Applied to Cochlear Partition on Discharges in Individual Auditory‐Nerve Fibers. II. Interaction of Electrical Polarization and Acoustic Stimulation

Donald C. Teas, Teruzo Konishi, and Joel S. Wernick

J. Acoust. Soc. Am. Volume 47, Issue 6B, pp. 1527-1537 (1970); (11 pages)

Online Publication Date: 11 Aug 2005

Full Text: | Download PDF

Show Abstract
This report describes the temporal patterns of discharges of single nerve fibers recorded by a microelectrode in the modiolus in response to short‐term dc (5‐sec) and ac stimulation delivered across scala media in the basal turn of the guinea pig's cochlea. These electrical stimuli produce increases or decreases in discharge rates, depending on the polarity of the current, but do not show adaptation effects when the current strength is moderate. Currents with scala vestibuli positive with respect to scala tympani (SV‐ST) are usually excitatory, while currents with ST‐SV produce reductions in discharge rate. The effect of a given polarity is found to depend on a unit's prestimulation spontaneous rate. When electrical stimulation is added to acoustic stimulation, the intensity function at best frequency is shifted laterally, with the direction of the shift depending on current polarity. When the effects of two sources of stimulation, acoustic and electric, are added, the maximum discharge rate seen for acoustic inputs is not altered; it only occurs at a lower strength of the acoustic signal. The addition of electrical stimulation to acoustic frequencies other than best frequency may show different effects.

Peripheral Inhibition in Auditory Fibers in the Frog

Harold J. Liff and Moise H. Goldstein, Jr.

J. Acoust. Soc. Am. Volume 47, Issue 6B, pp. 1538-1547 (1970); (10 pages)

Online Publication Date: 11 Aug 2005

Full Text: | Download PDF

Show Abstract
Two‐tone interactions in complex units in bullfrogs and leopard frogs were studied. The discharges of a complex unit in response to a tone at its best excitatory frequency (BEF) could be suppressed by addition of a second, higher‐frequency tone. The best inhibitory frequency (BIF) was that for which the least intense tone totally suppressed the response to a BEF tone. Between BEF and BIF, there was a frequency range in which tones had both excitatory and inhibitory effects. We call this range the “gray region.” During prolonged continuous presentation of a high‐level inhibiting tone, the inhibitory effect showed no adaptation. A rebound in firing rate occurred when a unit, which had been previously adapted and then inhibited, was released from the inhibition by terminating the inhibitory tone. The level of an inhibitory tone necessary for total inhibition depended on excitor level, even for levels at which the spike rate in response to the excitor alone had saturated. Results are discussed with regard to the stage in the process leading to neuronal excitation at which inhibition occurs, and two‐tone interactions in frog complex units are compared with those in mammalian first‐order units.

Monaural Detection and Filtering

B. E. Mulligan and M. Elrod

J. Acoust. Soc. Am. Volume 47, Issue 6B, pp. 1548-1556 (1970); (9 pages)

Online Publication Date: 11 Aug 2005

Full Text: | Download PDF

Show Abstract
Monaural detection of sinusoidal signals in noise is examined parametrically with an interest in extending our earlier work on prediction. The amplitude model is found to be compatible with empirical psychometric functions, human receiver operating characteristics, and the results of a narrow‐band‐noise experiment. Through an interpretation of the amplitude model in terms of filter bandwidth, an attempt is made to understand further the process of auditory filtering. Bandwidth is shown to vary as a function of both signal frequency and signal‐to‐noise (S/N) ratio. At very low S/N ratios, maximal tuning occurs in the region of 2200–2300 Hz. At higher S/N ratios, tuning improves as a function of frequency. To facilitate prediction, a table is made available. The table gives corresponding values of [(S/N0)⅓ − k], P[Hit], ds, (2S/N0W)½, and p[C].

Subjective and Objective Measurement of the Loudness Level of Single and Repeated Impulses

W. Reichardt

J. Acoust. Soc. Am. Volume 47, Issue 6B, pp. 1557-1562 (1970); (6 pages) | Cited 1 time

Online Publication Date: 11 Aug 2005

Full Text: | Download PDF

Show Abstract
The subjective measurement of impulsive sounds involves equal duration for the 1‐kHz comparison tones and the same repetition rate as the impulses being measured. The loudness levels of the short 1‐kHz tones and repeated short 1‐kHz tones can be taken from nomograms presented. The objective measurement of impulses requires an impulse sould‐level meter with an RC‐integrating network following the squaring device. The RC time constant must simulate the inertia of the human ear. The rise time and decay time constant must be the same. This impulse SLM sound‐level meter indicated the same effective value as the classical sound‐level meter for steady noises or for rapidly repeated impulses. The value indicated for slowly repeated impulses is higher, and we have a continuous transition to this higher value. The decay time constant must correspond to the rate of decrease in loudness after the sound has stopped. This can be described as τ = 25 msec or as 60‐dB reverberation time of 0.35 sec. The reading of such an impulse sound‐level meter with A weighting is designated LAI. Loudness calculation by Niese's (1956. Hochfrequenztech. u. Elektroakust. 65, 4.) method, for either single or repeated impulses, requires a spectrum correction ΔLSP to be taken from a nomogram, LNI = LAI+ΔLSP. Stevens′ (1961. J. Acoust. Soc. Amer. 33, 1957) and Zwicker′s (1960. Acustica 10, 304) methods, for single impulses or impulses repeated less frequently than 2/sec, require the dynamic impulse level LCI to be measured in 1‐oct or ⅓‐oct bands. Impulses repeated more frequently than 2/sec are measured and calculated in the classical way, but need by this method another impulse correction ΔLI = LAI−LAS.

Resistance‐Capacitance Oscillator for Generating the Frequencies of the Equally Tempered Musical Scale

D. L. H. Gibbings

J. Acoust. Soc. Am. Volume 47, Issue 6B, pp. 1563-1568 (1970); (6 pages)

Online Publication Date: 11 Aug 2005

Full Text: | Download PDF

Show Abstract
The frequencies of the equally tempered chromatic scale and the attenuation ratios of a uniform ladder attenuator are both in geometric progression. By modifying a Wien bridge oscillator so that the shunt resistance arm of the bridge is driven from a ladder attenuator, an oscillator has been devised that will generate the frequencies of the scale without requiring awkward component values. Two feedback amplifiers are used: one of them is frequency sensitive, and the other provides the necessary phase reversal and amplitude control. With simple transistor amplifiers, the frequency of oscillation follows the attenuator setting to within about 5 in 104, but ways are described by which the frequency of any semitone within an octave can be made correct to 1 or 2 in 104, without demanding comparable accuracy in the components. The circuit also allows the ratio between semitones to be progressively changed by the small amount required to accommodate the phenomenon of “octave stretching.” The design of an attenuator, using only preferred resistance values, is given, and the question of temperature coefficient and adjustment to standard frequency, A=440 Hz, is considered.

Chords from Tones Having Stretched Partials

Frank H. Slaymaker

J. Acoust. Soc. Am. Volume 47, Issue 6B, pp. 1569-1571 (1970); (3 pages) | Cited 5 times

Online Publication Date: 11 Aug 2005

Full Text: | Download PDF

Show Abstract
Tones made up of partials having frequencies proportional to NS, where N is the partial number and S is the stretch exponent, have been generated on a digital computer and used to form scales and chords. The scales were constructed with the frequency ratio between “half steps” equal to (21/12)S. Chords analogous to the tonic, subdominant, and dominant were generated for S varying from 0.792481, which compressed the partials by a factor that made the frequency of the fourth partial three times that of the fundamental, to 1.261859, which made the frequency of the third partial four times that of the first. The sensation produced by the stretched chords was consistent with Helmholtz's criterion for dissonance. The quality of the individual tones, however, became more and more strange as the value of S increased.

Formant‐Frequency Trajectories in Selected CVC‐Syllable Nuclei

David J. Broad and Ralph H. Fertig

J. Acoust. Soc. Am. Volume 47, Issue 6B, pp. 1572-1582 (1970); (11 pages) | Cited 4 times

Online Publication Date: 11 Aug 2005

Full Text: | Download PDF

Show Abstract
The frequencies of the first three vowel formants in 1728 monosyllables containing the Midwestern American English vowel /ɪ/ (as in fit) were measured by means of a computer‐assisted technique employing a data‐display screen, light pen, and television camera. Two‐way analyses of variance for various time points of the formant trajectories reveal that nonlinear interaction between initial and final consonants is relatively small and that the influences of both initial and final consonants are highly significant throughout the vowel. A linear model gives the mean formant trajectory for a given consonantal context as the sum of a target frequency, an intial consonant transition function, and a final consonant transition function. The distributions of the differences between the model trajectories and the actual trajectories are reasonably approximated by normal stochastic processes. These distributions are interpreted not as errors of the model, but, rather, as due predominantly to the variability between syllable tokens of the same syllable type.

Reduction of Observer Bias in Reading Speech Levels with a VU Meter

H. Levitt and P. D. Bricker

J. Acoust. Soc. Am. Volume 47, Issue 6B, pp. 1583-1587 (1970); (5 pages)

Online Publication Date: 11 Aug 2005

Full Text: | Download PDF

Show Abstract
A procedure is described for reading the levels of certain types of speech sources using a VU meter with minor modifications. The method involves random sampling of the waveform, a simple decision about pointer position by the observer, and strict rules for adjusting the range attenuator. The procedure is shown to reduce interobserver variability considerably when a substantial amount of speech is available from a relatively constant source.

The Lenis‐Fortis Opposition: Its Physiological Parameters

André Malécot

J. Acoust. Soc. Am. Volume 47, Issue 6B, pp. 1588-1592 (1970); (5 pages) | Cited 1 time

Online Publication Date: 11 Aug 2005

Full Text: | Download PDF

Show Abstract
Evidence reviewed in this article indicates that the lenis‐fortis feature of consonants is primarily a synesthetic interpretation of magnitudes of intrabuccal air pressure and is conveyed variously in different contexts by the durations of the consonant closure and of the preceding vowel. Variations in magnitude as a function of idiolect, syllabic rate, phonetic context, etc., are also described, and new data from normal speech are presented.

Performance Characteristics of an Experimental Harmonic Identification Pitch Extraction (HIPEX) System

R. L. Miller

J. Acoust. Soc. Am. Volume 47, Issue 6B, pp. 1593-1601 (1970); (9 pages)

Online Publication Date: 11 Aug 2005

Full Text: | Download PDF

Show Abstract
A description is given of the HIPEX (Harmonic Identification Pitch Extraction) pitch measurement system. This experimental system is based on finding the common denominator of harmonically related components as separated out by narrow‐band filters. The performance characteristics are given in relation to system‐parameter changes. A statistical study of “period histograms” with a wide range of pitches gives an optimum peak‐detection ratio of 0.91 to identify the peak corresponding to the fundamental period. Performance of the HIPEX system in the presence of noise is unusually good. Usable pitch information was obtained from speech in an approximate zero S/N condition. The effectively coherent addition of harmonic signals in the period histogram, regardless of their phase in the speech wave, implies that the technique is approaching an optimum for determination of the fundamental period from the available information.

Measurement System for Fetal Audiometry

R. J. Bench, J. H. Anderson, and Margaret Hoare

J. Acoust. Soc. Am. Volume 47, Issue 6B, pp. 1602-1606 (1970); (5 pages)

Online Publication Date: 11 Aug 2005

Full Text: | Download PDF

Show Abstract
There has recently been renewed interest in the early development of audition, especially in the fetus and neonate, but new techniques are required before reliable substantive findings can be made. A measurement system is described for assessing the capacity of the animal (goat) fetus for response to audio‐frequency stimulation. Particular emphasis placed on the characteristics and development of an internally implanted hydrophone for monitoring sound levels close to the fetal head. Sound levels generated by an externally mounted vibrator and monitored at the uterine wall underwent attenuation increasing with gestational age. This is attributed to changes in the geometry of the goat with advancing pregnancy.

The Role of Heat in the Production of Ultrasonic Focal Lesions

John B. Pond

J. Acoust. Soc. Am. Volume 47, Issue 6B, pp. 1607-1611 (1970); (5 pages) | Cited 2 times

Online Publication Date: 11 Aug 2005

Full Text: | Download PDF

Show Abstract
A method was evolved to calculate the temperature cycles at the focus center of an intermittent beam of high‐frequency mechanical waves. The computed temperatures were checked in brain tissue, using microthermocouples, for doses just sufficient to damage tissue at the focus center. Similar temperature cycles were produced in brain tissue, in the absence of mechanical wave energy, by applying pulses of electric current to embedded resistance wires in order to find out how far the tissue alteration could be ascribed to heating.
back to top
RSS Feeds

Implications of Speech‐Recognition Studies

Lawrence Fleming

J. Acoust. Soc. Am. Volume 47, Issue 6B, pp. 1612-1612 (1970); (1 page)

Online Publication Date: 11 Aug 2005

Full Text: | Download PDF

Show Abstract
Automation has reduced many human tasks to subhuman‐skill requirements. In this context, speech‐recognition machines appear to be a logical further step.

Evaluating Speech‐Recognition Work

Wayne A. Lea

J. Acoust. Soc. Am. Volume 47, Issue 6B, pp. 1612-1614 (1970); (3 pages)

Online Publication Date: 11 Aug 2005

Full Text: | Download PDF

Show Abstract
Pierce [J. Acoust. Soc. Amer. 46, 1049–1051(L) (1969)] has recently questioned motivations and progress in speech‐recognition work. Constructive steps can and should be taken to determine the value of speech communication with computers, the potential for “success,” the value of limited speech recognizers, and the best design of speech‐research programs. Speech does offer some advantages for effective man‐computer interaction. Since fully versatile speech recognizers are apparently not immediately forthcoming, scientific studies should be made of the actual utility of restricted recognizers. One such study has begun. Pierce's criticisms remind one of the continued need to answer elementary questions of “why” and “how much” in speech‐recognition work.

Whither Speech Production?

A. Michael Noll

J. Acoust. Soc. Am. Volume 47, Issue 6B, pp. 1614-1616 (1970); (3 pages)

Online Publication Date: 11 Aug 2005

Full Text: | Download PDF

Show Abstract
Research in automatic speech production, like research in automatic speech recognition, seems to be regressing at an ever‐increasing rate when compared with the over‐all progress of technology. Both speech production and speech recognition involve human beings and perceptual phenomena about which very little information is available. A negative attitude towards either automatic speech recognition or automatic speech production logically must lead to a negative attitude towards the other. Hence, speech research aimed at automating either the recognition or the production of speech will require many years of effort before anything approaching practical results is obtained. The speech process should ultimately be investigated for its own intrinsic merit—not for satisfying man‐machine communication needs.

“Whither Speech Recognition?”—A Rebuttal

A. L. Samuel

J. Acoust. Soc. Am. Volume 47, Issue 6B, pp. 1616-1616 (1970); (1 page)

Online Publication Date: 11 Aug 2005

Full Text: | Download PDF

Show Abstract
There are honest scientific workers in the field of speech recognition who are not blinded by glamor and deceit, who do not behave “like mad inventors or untrustworthy engineers” and who have honest answers to the questions raised by John Pierce [“Whither Speech Recognition?,” J. Acoust. Soc. Amer. 46, 1049–1051(L) (1969)].

Whither Speech Recognition?—II

J. R. Pierce

J. Acoust. Soc. Am. Volume 47, Issue 6B, pp. 1616-1617 (1970); (2 pages)

Online Publication Date: 11 Aug 2005

Full Text: | Download PDF

Show Abstract
Mr. Pierce does not recant.

Paced Recognition of Words Masked in White Noise

C. M. Holloway

J. Acoust. Soc. Am. Volume 47, Issue 6B, pp. 1617-1618 (1970); (2 pages)

Online Publication Date: 11 Aug 2005

Full Text: | Download PDF

Show Abstract
An experiment is reported in which listeners make recognition responses to words mixed with one of five levels of noise and presented at one of five rates of presentation. An interaction is demonstrated between the effects of noise level and rate of presentation, which cannot be attributed to differences in enunciation of the test material at different rates of presentation.
Close

Home Publications Meetings Contact ASA Site Map Back to Top

Copyright © Acoustical Society of America

close