摘要:

A flexible, intracochlear multimicroelectrode array is being developed for differential electrical stimulation of the cochlear nerve in human sensory deafness [M. Sonn, Proc. 25th Annual Conf. Eng. Med. Biol.14, 159 (1972)]. The conductor elements, 0.3 μm thick, are formed on both sides of a thin‐film plastic substrate using vacuum deposition and photolithographic etching techniques. In the present configuration, the array consists of 33 platinum conductors, 12.7 μm wide, each terminating at an electrode surface, 250 μm wide. The electrodes are spaced on 700 μm centers. Substrates investigated include Mylar, Parylene‐C, and FEP and TFE Teflon films, 50 to 130 μm thick. Of these, Mylar was found best for electrode fabrication. Insulation of the array is accomplished by dip coating in Silastic. The electrode tips are exposed by mechanical scribing techniques. Short‐term body tolerance studies in guinea pigs show good compatibility with the electrode materials. Bipolar sinusoidal stimulation of the electrodes in normal saline shows a decrease in complex impedance proportional to the −0.99 power of the frequency: 23.8 MΩ at 100 Hz to 250 kΩ at 10 kHz. Stimulation studiesin vitroandin vivo(scalae tympani of cats) are planned to determine behavioral responses to electrical stimuli and long‐term tissue reaction and electrolytic dissolution effects. [This work was supported in part by Raytheon Company.]

ISSN:0001-4966    

DOI:10.1121/1.1982343

出版商:Acoustical Society of America    

年代:1973

数据来源: AIP

摘要:

The channel capacity of the ear as a function of frequency can be computed by a technique similar to that in an earlier paper [E. L. R. Corliss, J. Acoust. Soc. Amer.50, 671–677 (1971)]. The legitimacy of this computation rests upon the justification that sufficient variables are involved to permit replacement of actual statistical distributions by normal distributions. This legitimacy is tested by comparing the predicted growth of intelligibility of individual phonemes as a function of level above threshold with experimental results. The channel capacity of the ear shows a broad maximum in the frequency range of 3–6 kHz. Taking into account the duration of speech sounds as a function of frequency, one finds that each phoneme tends to convey the same number of bits over the range from 300 Hz to 4 kHz. However, many more phonemes are uttered per unit time at the higher frequencies, so that the rate of information transfer is higher. This explains to some extent the empirical observation that the high‐frequency limit necessary for accurate communication depends upon redundancy of the material. One ought to reconsider the convention that the spectrum above 3 kHz is relatively irrelevant for speech communication when unfamiliar speech material must be recognized.

ISSN:0001-4966    

DOI:10.1121/1.1982346

出版商:Acoustical Society of America    

年代:1973

数据来源: AIP

摘要:

The acoustic composition of voiced speech signals received at a distance of 1 m (normal speech) is compared with those picked up by standard Air Force lip microphones that are transmitted via standard electroacoustic communication systems. The acoustic composition of speech used in Air Force ground and airborne operations is significantly different from that associated with normal live voice communication. Many voiced and voiceless speech components that constitute minimal acoustic amplitude in normal speaking situations undergo significant amplitude alterations when picked up by lip microphones. The author noted that pilots and other rated personnel frequently report that several transition flights are required before the speech communication messages are fully intelligible. This phenomenon has been studied and analyses of actual airborne speech communications are used to identify and define acoustic alterations that necessitate learning a new (or altered) type of speech. Voice prints of English vowels, diphthongs, and consonants are used to illustrate acoustic changes that exist between normal conversational speech and speech picked up by lip microphones. The results of this research suggest that the acoustic alterations created by picking up speech at the lips may increase intelligiblilty after the listener has learned how to process the signals accurately.

ISSN:0001-4966    

DOI:10.1121/1.1982353

出版商:Acoustical Society of America    

年代:1973

数据来源: AIP

摘要:

In this preliminary study, we attempt to relate the space defined by a statistical [principal component] analysis of speech spectra to perceptual space represented by intelligibility scores. Speech spectra were represented in terms of a varying number of dimensions [3, 5, 7, 10, 15, and 35], each corresponding to a different degree of statistical accuracy, and the waveform was reconstituted from the reduced spectral data by means of inverse FFT computation [K.‐P. Li, G. W. Hughes, and A. S. House, J. Acoust. Soc. Amer.49, 134(A) (1971)]. Consonantal intelligibility scores for isolated monosyllabic utterances by one talker were determined from the responses of 12 listeners to a total of 553 test items. The average score increases monotonically from 47% to 75% as the spectral dimensionality varies from 3 to 35. Along with the number of dimensions, manner and place of consonant articulation and the effects of the details of the reconstitution process enter into the relationship between statistical accuracy and intelligibility.

ISSN:0001-4966    

DOI:10.1121/1.1982358

出版商:Acoustical Society of America    

年代:1973

数据来源: AIP

摘要:

Proposed lift‐augmented STOL aircraft are expected to require noise reductions on the order of 20 dB to meet community noise goals, according to a number of recent model studies. A predominant source of noise associated with the concepts and configurations currently under consideration is the interaction of the exhaust stream with deployed flaps. This flow‐surface interaction produces both large‐scale aerodynamic force fluctuations, which radiate like point dipoles at low frequencies, and smaller‐scale surface pressure fluctuations, which radiate only from the leading and trailing edges of these surfaces. The fundamentals of the relevant noise‐producing processes are reviewed and basic noise reduction concepts developed using parametric source models. Experimentally observed noise reductions on a 1/15 scale externally blown flap (EBF) system are presented for two of the noise reduction concepts—modification of the acoustic transduction process using variable impedance surfaces, and modification of the incident flow field through introduction of high porosity mesh grids upstream of the flaps.

ISSN:0001-4966    

DOI:10.1121/1.1982362

出版商:Acoustical Society of America    

年代:1973

数据来源: AIP

摘要:

Some results of wind‐tunnel studies of helicopter noise carried out on a model rotor operating in a large open‐jet wind tunnel are described. The tunnel test section is surrounded by an anechoic chamber. Data processing techniques were used to obtain the average‐repeated‐transient signal per revolution and thus to extract the rotational noise from the total noise, which includes random components. Blade‐vortex interaction is clearly visible in many of the repeated‐transient signals and is felt to be an important contributor to rotational noise even outside of the brade‐slap condition. Vortex noise can be identified in the spectra of the total signal. In addition to the vortex noise, a broad‐band noise of lower frequency is evident. It is postulated that this lower‐frequency noise is due either to vortex‐wave unsteadiness or to local stall on the blades at certain azimuth locations.

ISSN:0001-4966    

DOI:10.1121/1.1982364

出版商:Acoustical Society of America    

年代:1973

数据来源: AIP

摘要:

The surface duct is the mixed‐layer underwater sound channel generally bounded by the ocean surface and the thermocline. In a recent experiment, abundant measurements of surface‐duct propagation loss were made concurrently with detailed environmental sampling. The experiment requiring three ships was designed to fill a gap in the knowledge of surface‐duct propagation at frequencies below 8 kHz that has not been addressed. Propagation loss was measured at frequencies from 0.4 to 20.0 kHz, at depths from 20 to 600 ft, over ranges of 0–18 miles. The experiment was novel because for the first time adequate environmental measurements were made over the entire acoustic field. Three different instruments measured sea surface roughness. Sound speed profiles were continuously sampled both in space and time: every 50 ft along the track by a towed thermistor chain, every 10 sec by a string of thermistor buoys spaced along the propagation path. The accuracy of these systems was ±0.02°C or ±0.06 m/sec in terms of sound speed. The purpose of the experiment was to provide data required for evaluating several theories of surface duct propagation loss.

ISSN:0001-4966    

DOI:10.1121/1.1982372

出版商:Acoustical Society of America    

年代:1973

数据来源: AIP

摘要:

Results from the studies described in the two preceding abstracts have been combined in models that describe the effects of internal tides and surface tides upon multipath acoustic propagation. Deep ocean experimental data that motivated this analysis will be discussed briefly. Several examples of ray multipath propagation in deep ocean channels time‐varying under the influence of internal and surface tides will be examined. Comparisons will be drawn with analogous results from shallow water studies.

ISSN:0001-4966    

DOI:10.1121/1.1982377

出版商:Acoustical Society of America    

年代:1973

数据来源: AIP

摘要:

Loudness summation between two equally loud, 500‐msec tones centered on 2000 Hz was measured in an anechoic room as a function of the frequency separation, ΔF, and onset time difference, ΔT, between the tones. One tone was presented via a loudspeaker 3 m from the subject and 45° to the left of his median plane, the other tone via a loudspeaker 45° to the right. A comparison 2000‐Hz tone was presented via a third speaker in the median plane. Overall loudness increased with ΔFonce the critical band was exceeded. When ΔTwas 0 msec, loudness summation at 45, 65, and 85 phons was very similar to that between two tones presented through a single earphone. As ΔTincreased from 0 to about 100 msec, loudness summation declined gradually; beyond 100 msec, it declined more rapidly. The Haas or precedence effect was evident at ΔT's less than 40 to 60 msec and ΔF's less than about 1000 Hz, where the six subjects usually heard a single sound originating toward the side of the leading loudspeaker.

ISSN:0001-4966    

DOI:10.1121/1.1982379

出版商:Acoustical Society of America    

年代:1973

数据来源: AIP

摘要:

The precision of a human underwater sound localization response was defined as the correspondence between the objective azimuth of the sound source and the azimuth indicated by a pointer controller by the subject. It was determined that the pointing response had an average error of 5.98 when the subjects pointed to a visual target located in the frontal plane. When the subjects were required to locate an acoustic target located at some point in 360°, the average error was 19.2°. Front to back (or back to front) reversals occurred at 0° and 180°±45° and amounted to 15.1% of the total responses. A replication of the visual target localization in 360° yielded an average error of 6.98° After 240 training trials in which the subject was allowed to visually correct his auditory localization, the average error in 360° was 13.1° Reversals were reduced to 7.1% of the total responses.

ISSN:0001-4966    

DOI:10.1121/1.1982390

出版商:Acoustical Society of America    

年代:1973

数据来源: AIP