摘要:

When a vehicle is set in motion, its flow noise level increases with the speed. If the surface is rough, flow noise level increases with speed and without any noticeable time lag. If the surface is highly polished, it takes a few seconds for the flow noise to build up. As a consequence, the level of the flow noise and its spectrum are dependent upon the history of the flow before the flow reaches the sound receiver. The amplitude distribution of the flow noise turns out to be very nearly Gaussian. However, in certain instances flow noise is emitted in bursts. Experimental results will be presented.

ISSN:0001-4966    

DOI:10.1121/1.1935172

出版商:Acoustical Society of America    

年代:1960

数据来源: AIP

摘要:

Relative sound output of bar magnetostrictors has been measured as a function of polarization at atmospheric stress and at a compressional stress of about 300 atm. Frequencies both at and well below the resonant frequency of the bar have been used. Thin laminations of nickel or permendur are glued together to form bars whose ends are then ground parallel. Each bar is length resonant at about 23 kcps. Half‐wavelength brass stubs with large flanges at their centers are soldered to the bar and loosely mounted in cylinders. Air can be forced between the flanges and cylinder walls to apply compressive stress to the bar. The displacement of the end of one of the stubs is observed. For nickel, compressive stress increases sound output at low polarizations and decreases sound output at higher polarizations. Representative values are these: a compressive stress of 300 atm increases the resonant amplitude by 15% at a polarizing field of 20 oe and decreases it 15% at 70 oe. The polarization at which 300 atm compressive stress does not alter the resonant displacement is 45.3 oe. Displacements at 60 cps and at resonance are related by a theory involving the magnetostrictive constant and the reversible permeability, and there is good agreement between this theory and experiments. (This research was supported by the Office of Naval Research.)

ISSN:0001-4966    

DOI:10.1121/1.1935176

出版商:Acoustical Society of America    

年代:1960

数据来源: AIP

摘要:

This paper describes a Progressive Wave System, 1 in. × 1 in. in cross section, employing one electro‐pneumatic transducer in the frequency range 37.5–1200 cps. The unit is capable of generating sound pressure levels in excess of 160 db over‐all sound pressure level. High‐frequency energy is generated by conventional compression‐type dynamic transducers. The high‐frequency energy is combined with the low‐frequency energy in a mixing unit which consists essentially of a 45° acoustic mirror. By means of appropriate slots, the mirror is made transparent in the low‐frequency range and reflective at the higher frequencies. Design data on the acoustic mirror and measurements of over‐all and octave band sound pressure levels, as well as variations in sound pressure over the cross section of the tube, particularly in the transition range from the plane to the progressive wave condition, are presented. The capability of the system when four electro‐pneumatic transducers are employed also is discussed.

ISSN:0001-4966    

DOI:10.1121/1.1935183

出版商:Acoustical Society of America    

年代:1960

数据来源: AIP

摘要:

It is generally accepted that the voiced portions of speech may be idealized into time variations of the first three formant frequencies and the fundamental frequency. The influence of these parameters upon the classification of short steady‐state vowel segments was determined. The mechanism by which a listener quantizes the voiced portions of speech into units subjectively equal to the steady‐state vowel segments is presented. The application of this method of vowel classification to an automatic speech recognition process is discussed.

ISSN:0001-4966    

DOI:10.1121/1.1935190

出版商:Acoustical Society of America    

年代:1960

数据来源: AIP

摘要:

A study has been made of the influence of transient and intermodulation distortion on the transmission of speech occurring in the presence of a background noise having pronounced transient characteristics (cafeteria noise). This paper describes objective measurements of the power‐time distribution in speech and noise, separately and in combination, for the sound coming from a sound source of high quality and for the sound coming from the earphones of hearing aids whose microphones were exposed to the sound source. The nature of the transmitted signal will be demonstrated. (This work was carried out under sponsorship of Veterans Administration.)

ISSN:0001-4966    

DOI:10.1121/1.1935194

出版商:Acoustical Society of America    

年代:1960

数据来源: AIP

摘要:

Pulses of ultrasound of finite amplitude are generated in liquids by an electrical system of 2.5‐Mc frequency and an air‐backed x‐cut quartz crystal of the same fundamental frequency. After passage through the liquid, the signals are received on a 2‐in.diameterx‐cut quartz crystal with a fundamental frequency of 25 Mc. The distortion is measured either by a Fourier analysis of the received waveform or by measurement of the various harmonics by a tunable amplifier. Photographs of the waveform at different distances from the transducer illustrate the formation of the sawtooth wave. Comparison is made between the measured harmonic content and that computed theoretically. [The work was supported in part by the Office of Naval Research and by a National Research Council Fellowship to one of the authors (A.G.L.).]

ISSN:0001-4966    

DOI:10.1121/1.1935202

出版商:Acoustical Society of America    

年代:1960

数据来源: AIP

摘要:

Earthquake waves as they proceed over the earth's surface radiate sound waves into the atmosphere. Several instances of such sound waves have been observed at Washington, D. C. A notable instance was the great Montana earthquake of August 18, 1959 (UT). Calibrated microphones responding to very low frequencies measured the amount of the radiated sound pressure and the time delays between sound pressures at several geographic locations. From these have been deduced the vertical amplitudes of the seismic waves and their speed of transit through the Washington area. Recordings of the waves will be presented. Results of the acoustical measurements will be compared with seismological observations. With respect to one of the Chilean earthquakes, sound radiated at the epicenter was apparently propagated through the atmosphere to Washington.

ISSN:0001-4966    

DOI:10.1121/1.1935203

出版商:Acoustical Society of America    

年代:1960

数据来源: AIP

摘要:

Two fixed‐interval observation experiments were conducted to determine the effect ofa prioriprobability of signal occurrence [p(SN)] on signal detectability. The signal was a “pure tone” of 1000 cps, presented for 250 msec against a continuous background of white gaussian noise. (1) Three values ofp(SN) (0.25, 0.50, and 0.75) were paired with each of three values ofE/N0(6.3, 12.6, and 20.0). (2) WithE/N0fixed at 12.6,p(SN) was assigned values of 0.10, 0.25, 0.50, 0.75, and 0.90. Listeners employed a four‐point rating scale to indicate their confidence that a signal had been presented. Operating characteristics were fitted to the data on normal probability paper. An obtained operating characteristic indicates the detectability of the signal for the listener and may be specified completely by its two parameters, slope and √de. The results indicate that √deis independent ofp(SN), as predicted by the Theory of Signal Detectability. The slope of the operating characteristic, on the other hand, appears to be a joint function ofp(SN) andE/N0. WithE/N0= 20.0, slope increased markedly asp(SN) increased; withE/N0= 12.6, slope was a nondecreasing function ofp(SN), ranging from about 0.70 forp(SN) = 0.10 to about 1.00 forp(SN) = 0.90; and withE/N0= 6.3, slope was independent ofp(SN). (This research was supported by the Operational Applications Laboratory, Air Force Cambridge Research Center.)

ISSN:0001-4966    

DOI:10.1121/1.1935214

出版商:Acoustical Society of America    

年代:1960

数据来源: AIP

摘要:

Gated random noise was used as a stimulus in a psychophysical experiment to determine whether a timing signal was present for certain conditions of gating frequency, gating duration, and noise spectrum. The stimulus was presented via earphones as all pass or 2400 cps high pass. The subjects controlled the gating frequency matching it to a 0.1‐msec click, and the standard deviation of the match was measured. All stimuli had sensation levels near 40 db. Ungated noise was added to the gated noise to mask any switching transients. For aDF(% of time noise was gated on) of 20% and 50%, some timing signal in the noise could be matched to a click of 750 cps. All‐pass clicks were easier to match than high‐pass clicks. For aDFof 80%, there was no timing signal above a frequency of 200 cps. This also checked that there was little information given to the subject through the frequency controlling dial. In an auxiliary experiment subjects could fuse gated, random, binaurally incoherent, noise on some but not all of the conditions for which a pitch match was obtained.

ISSN:0001-4966    

DOI:10.1121/1.1935219

出版商:Acoustical Society of America    

年代:1960

数据来源: AIP

摘要:

The space‐time decision theory of acoustic fields is developed from an examination of the conditions necessary for uniform and nonuniform spatial sampling. Basic transformation relations between spatial sampling and direction sampling are given. From this sampling theory, the monochromatic space likelihood ratio, the broadband space likelihood ratio, and the general space‐time likelihood ratio are developed. Examples of the general theory include the point, blurred point, and ring signal fields and several (spatially) statistical noise fields. Superdirectivity and the spatial matched filter are related to the space‐time likelihood ratio, and the paper concludes with a brief discussion of necessary and complete information inherent in an acoustic field.

ISSN:0001-4966    

DOI:10.1121/1.1935224

出版商:Acoustical Society of America    

年代:1960

数据来源: AIP