摘要:

Although the oscillation frequency of a flue organ pipe can usually be measured within 1% or less, observations of the signal with an oscilloscope may show small, apparently random fluctuations in the waveform. These fluctuations have been investigated with a model pipe having a specially designed flue to discharge turbulence below the jet opening. Nonperiodic fluctuations in the fundamental amplitude are found by digital signal processing. For example, the fluctuations are 1–2% for these conditions: Resonator section, 38 × 57 mm; length, 360 mm; lip height, 6.5 mm; jet opening, 0.6 mm; lip radius 0.2 mm; and supply pressure 45‐mm water. The instability is attributed to interaction between the resonator feedback mechanism and the edge‐tone feedback mechanism. (The tone has a “sizzle” which is commonly regarded as a vestigal edgetone effect.) The “partials” near the edgetone frequency (about four times the speaking frequency in this case) are shown to be effectively noise bands. Successive amplitude samples from this region have significant cross correlation with those from the fundamental.

ISSN:0001-4966    

DOI:10.1121/1.2017018

出版商:Acoustical Society of America    

年代:1979

数据来源: AIP

摘要:

Personal noise dosimetry monitoring devices have been used for some time by various Air Force bases to determine the occupational noise exposures of Air Force personnel. In May 1978, Aerospace Medical Research Laboratory began consolidating noise dosimetry data from these bases. The author has personally collected data during the past year on visits to four Air Force bases. Over 418 separate 8‐h occupational noise exposure measurements were made on 120 different people. Some of the research involved 16 people wearing a dosimeter for 1 day or up to a week at a time. Seventy‐three separate 24‐h measurements were made. From these data a comparison was made between the off‐duty noise exposures and the work‐related noise exposures. The individual's work‐related noise exposure (except in a few instances) contributed the highest noise levels to the person's 24‐h noise dose. The 24‐h noise levels ranged from 60 to 92 dBAwith a median level of 77 dBA. Previous studies supported by AMRL have shown a median level of 75 dBAfor the typical American. This paper will discuss the basis of the AMRL research, as well as the merits of noise dosimetry and some of the problems encountered during the study. [Work supported by EPA and USAF.]

ISSN:0001-4966    

DOI:10.1121/1.2017024

出版商:Acoustical Society of America    

年代:1979

数据来源: AIP

摘要:

The flexural disk transducer has been recognized for many years as a viable candidate for applications requiring low‐frequency sound sources. Since the establishment of the theory by Woollett [“Theory of the Piezoelectric Flexural Disk Transducer with Applications to Underwater Sound,” USL Report No. 490, December 1960], much of the development effort has been directed toward improved fabrication and mounting techniques which have resulted in high electroacoustic efficiencies, increased power handling capabilities and greater correlation between theory and measurements. Programs to provide greater depth capability and improved bandwidths are continuing. The combination of specific requirements impose certain limitations upon the design of the flexural disk and dictate tradeoffs where possible. These limitations and tradeoffs are described as they relate to transducer applications at low frequencies requiring high acoustic‐source levels, broadband widths, and deep submergence operation. Two transducers have been fabricated, one below 100 Hz to demonstrate potential VLF applications and limitations, and another at 250 Hz for evaluation of the power capabilities of large diameter (33 in.) mosaic disks. The designs, test results, and a summary of possible design trade‐offs are presented.

ISSN:0001-4966    

DOI:10.1121/1.2017027

出版商:Acoustical Society of America    

年代:1979

数据来源: AIP

摘要:

The design criteria for a deep‐submergence source utilizing the acoustic principle of the Helmholtz resonator is discussed. For frequencies below 500 Hz, the size of any acoustic source is necessarily large if any great amount of power is to be radiated. The advantage of the Helmholtz resonator becomes apparent when compared to a typical Tonpiltz resonator. Disadvantages are highly reactive electrical loads, which require high driving voltage and reactive power. Trade‐offs between physical size and bandwidth can be made for maximum efficiency.

ISSN:0001-4966    

DOI:10.1121/1.2017030

出版商:Acoustical Society of America    

年代:1979

数据来源: AIP

摘要:

Approximations for the frequency of the new resonance of a small vapor bubble associated with evaporation and condensation are derived from the equations of “Vapor bubble dynamics” [R. D. Rinch and E. A. Neppiras, J. Acoust. Soc. Am.53, 1402–1410 (1973)]. Previous predictions of the new resonance frequency ω, were obtained by numerically solving a secular equation. The purpose of the approximations is to elucidate the significant parameters; for example, the surface tension σ is predicted to greatly influence ω. When the liquid's thermal conductivitykis sufficiently large, ω is proportional to σ2/3R−8/3, where R is the equilibrium radius of the bubble; smallkgive ω proportional to σ2R−4. The approximations are substantiated by comparisons with numerical results for water, nitrogen, and helium. Criteria for their use depend on the relative magnitudes of the thermal diffusion length of the liquid, other liquid parameters, andR. Effects of evaporation‐condensation resonance on radiation pressure and extinction of sound are suggested. Observed stable bubbles in helium [P. L. Marston and D. B. Greene, J. Acoust. Soc. Am.64, 319–321 (1978)] are larger than those predicted for resonance. [Work partially performed at Yale University and supported by ONR.]

ISSN:0001-4966    

DOI:10.1121/1.2017037

出版商:Acoustical Society of America    

年代:1979

数据来源: AIP

摘要:

Deformations of drops and bubbles opposed by surface tension and driven by radiation stresses at the interface are calculated using spherical harmonic expansions for the radial and tangential stresses. Superposed acoustic waves produce stresses which oscillate at the difference frequency ω of the waves in addition to static stresses. When the effects of viscosity on the acoustic waves are omitted. the tangential radiation stress vanishes; a procedure is proposed for calculating the radial stresses from the theory for “Acoustic Radiation Pressure on a Compressible Sphere” [K. Yosioka and Y. Kawasima, Acustica5, 167–173 (1955)]. The calculation of the response assumes incompressible second‐order flow and omits the body forces which are normally associated with acoustic streaming. Resonance phase shifts and enhancements of the response should occur when ω is close to the natural oscillation frequency of a mode. Quadrupole resonance phase shifts and enhancements have been observed [P. L. Marston and R. E. Apfel, J. Acoust. Soc. Am.63, S41(A) (1978): J. Colloid Interface Sci.68, 280–286 (1979)]. [Work performed at Yale University and supported by ONR.]

ISSN:0001-4966    

DOI:10.1121/1.2017039

出版商:Acoustical Society of America    

年代:1979

数据来源: AIP

摘要:

English syllables can be treated as a syllable core, decomposable into initial and final demisyllables, plus optional word‐final phonetic affixes (e.g., /s,z,t,d,θ/) [Lovins and Fujimura, J. Acoust. Soc. Am.60, S75 (A) (1976)]. An inventory of 835 demisyllables, five phonetic affixes, and 100 reduced‐vowel units (CV or VC syllables) has been prepared in the form of LPC parameters. This inventory allows us to generate any of the ten thousand possible English syllables. An initial demisyllable is composed of an initial consonant or cluster, plus a fixed‐length segment which includes the CV transition; a final demisyllable contains the major portion of the syllable nucleus, plus postnuclear consonant(s). We have found it generally appropriate to produce “tense” or “long” syllabic nuclei by concatenations such as /Cɪ−/ + /−iyC/ For stressed syllables, this allows a reduction of the set of initial demisyllable vowels to /ɪ,ɛ,æ,a,ʌ,ɔ,ᴜ/. Final demisyllables distinguish all fifteen nuclei. Only two spectral smoothing rules are needed to derive syllable cores from concatenated initial and final demisyllables. Results of naive listeners' perceptual evaluation of a sample set of concatenated monosyllables will be discussed. We will also present a tape recording of sentences produced from the inventory.

ISSN:0001-4966    

DOI:10.1121/1.2017054

出版商:Acoustical Society of America    

年代:1979

数据来源: AIP

摘要:

In an earlier report [C. A. McGonegalet al., J. Acoust. Soc. Am.62, S4(A) (Fall 1977).] the effects of several transmission systems on speaker verification by human listeners were investigated. There it was shown that the transmission system played a significant role in the speaker verification process. In this talk we show the effects of the transmission system on an existing automatic speaker verification system [A. E. Rosenberg, Bell Systems Technical Journal55, 723–744 (1976)] in which the measured features are pitch and gain as a function of time for a specified utterance. There were ten male and ten female customers and 40 male and 40 female imposters in this experiment. All utterances were recorded using a conventional telephone connection. Post processing included ADPCM coding and LPC vocoding. Preliminary results indicate that, on the average, the transmission system affects the verification accuracy, especially when the reference and test utterances are subjected to different transmission conditions.

ISSN:0001-4966    

DOI:10.1121/1.2017057

出版商:Acoustical Society of America    

年代:1979

数据来源: AIP

摘要:

At present, most of the applications at Dialog Systems involve numeric input of one kind or another. Applications accomodate telephone speech from a large population and range from telephone toll management and switching to telephone bill paying which involves rather lengthy data entry. All entry is by isolated word. This paper will concentrate on users of these systems. Training and user format design along with human and system performance will be discussed in the context of a system that has been in operation for about nine months with some 4000 users. Since the user is affected by constraints of voice recognition protocol and parameters of the voice recognition system, these topics will also be addressed. Particular emphasis will be given to two styles of digit input. In the first, each digit is verified on input. In the second, short strings of digits are input and then verified by the user. Data from about 60 subjects will be presented and trade‐offs between the two systems will be discussed. Dependent upon recognition rate, a mixture of the two styles gives an optimum solution.

ISSN:0001-4966    

DOI:10.1121/1.2017063

出版商:Acoustical Society of America    

年代:1979

数据来源: AIP

摘要:

Discrimination tests were made on persons with hearing impairments and on patients with multiple sclerosis. Results include estimates of horizontal minimum audible angle (MAA) at eight azimuths, vertical MAA straight ahead, and just‐noticeable differences in interaural time delay (time JND) and interaural amplitude ratio (amplitude JND). The standard stimulus was broadband (0.25–10 kHz), pulsed (1 s) noise. Conductive cases (>35‐dB loss) gave abnormal values in almost all tests. Symmetric sensori‐neural cases with speech discrimination scores above 90% gave roughly normal values in all tests, while the corresponding group with scores below 80% gave elevated values for the vertical MAA and the horizontal MAA on the sides. Meniere's cases had normal MAAs and time JNDs and an amplitude JND that was normal only at high levels. Neurinoma cases gave abnormal values in at least one measurement each, with large intersubject variability. Persons with only one functional ear showed no ability to discriminate interaural parameters but some gave MAAs within the normal range. For multiple sclerosis patients, time JNDs, amplitude JNDs, and vertical MAAs were affected independently. [Work supported by NIH and Hopital Cantonal, Geneva, Switzerland.]

ISSN:0001-4966    

DOI:10.1121/1.2017070

出版商:Acoustical Society of America    

年代:1979

数据来源: AIP