摘要:

A research tool is described for studying the effects of glottal‐pulse waveform on the speech signal. The system is based on a line analog synthesizer program which uses an on‐line glottal‐pulse‐waveform editor. This system allows specification of an arbitrary waveform, which is then taken as the source impulse to synthesize up to 1.5 sec of speech. This new tool has been used for generation of stimuli for listening experiments on glottal source effects.

ISSN:0001-4966    

DOI:10.1121/1.1977462

出版商:Acoustical Society of America    

年代:1971

数据来源: AIP

摘要:

In order to facilitate a direct comparison of recent models of pitch and loudness memory, identical recognition‐memory experiments were performed with pitch and loudness as the memory variables. A trial consisted of a roving standard, an interpolated interval of variable duration filled with a variable number of tones, a comparison tone equal to, higher, or lower than the standard, and a same‐different decision on the part of the subject. For both pitch and loudness, criterion and performance measures were related to the position of the comparison tone relative to the standard and to the position of the standard relative to the set of standards. The loudness‐memory model proposed by Durlach and Braida leads to a more adequate explanation of these results than the pitch‐memory models proposed by Massaro or Wickelgren.

ISSN:0001-4966    

DOI:10.1121/1.1977472

出版商:Acoustical Society of America    

年代:1971

数据来源: AIP

摘要:

For a number of years, we have investigated auditory function with physiological and psychological measures and with electronic models. We have come to the conclusion that the auditory system must carry out its analysis in the dimension of time rather than the dimension of frequency. There is a simple relationship between frequency and time in signals composed of pure tones and simple combinations of sinusoids, so when these signals are used to test the auditory system the results may be interpreted as indicative of either time or frequency analysis. In complex stimuli composed of transients, such as speech signals, the simple relationship between frequency and time no longer exists. The auditory response to these signals can only be meaningfully interpreted in the dimension of time. Information in the primary auditory neurons is encoded inPulse Density Modulation, a form not used by current engineering technology. The auditory system appears to be an analyzer ideally designed for processing brief transients.

ISSN:0001-4966    

DOI:10.1121/1.1977477

出版商:Acoustical Society of America    

年代:1971

数据来源: AIP

摘要:

Recent emphasis placed on adoption ofA‐weighted sound‐level measurements for identifying degrees of auditory risk has prompted investigations which facilitate and clarify compromises associated with the introduction ofA‐weighted levels. The authors have compared recent estimates of auditory risk which employA‐weighted measures with those utilizing octave‐band data. The results derived from this study yieldedC‐Avalues that can be used to equateA‐weighted measurements with octave‐band auditory‐risk criteria. Auditory‐risk contours are reported for protected and unprotected exposures measured withC‐ andA‐weighted sound levels. These contours also reportC‐Avalues from which equivalentA‐weighted levels for conditions of attenuation (earplugs, earmuffs, and communication headsets) can be determined for durations of exposure ranging from 2 to 480 min per day.

ISSN:0001-4966    

DOI:10.1121/1.1977484

出版商:Acoustical Society of America    

年代:1971

数据来源: AIP

摘要:

High‐power ultrasonics has been a useful tool in studying fatigue in metals, since a large number of cycles can be obtained in a small time. In the case of brass and iron, the mechanisms appear to be different, since at the high frequencies fatigue occurs in isolated regions, whereas at the low frequencies the slip is generally spread over the grains. The same mechanism can be produced at high frequencies by going to higher temperatures. In iron, the stress required for fatigue at higher frequencies is considerably less than for low frequencies. Recent measurements in titanium have shown that the required stress is about the same for high and low frequencies. The crack propagation velocity, however, is much faster at the ultrasonic frequencies. This was proved by striation measurements with a scanning electron microscope. A striation reflects the fact that the crack propagates under a different stress on one‐half of the cycle than it does on the other half. Hence, by counting the number of striations, the velocity can be calibrated; thus the high‐frequency fracture is not only localized, it is virtually catastropic in the localized area. [Supported by the Fatigue and Structural Reliability Institute under a contract with the U. S. Navy Office of Naval Research.]

ISSN:0001-4966    

DOI:10.1121/1.1977487

出版商:Acoustical Society of America    

年代:1971

数据来源: AIP

摘要:

Deformation of the surface of a soft solid by an ultrasonically vibrating object (20 kHz) was studied by means of a displacement transducer. An ultrasonic horn with a suitably formed tip was vibrated at amplitudes of 1 to 25μ against agar gel, wax, plastics, and soft metals while penetration into the test sample was recorded as a function of time. As an interesting example, a hemispherically shaped vibrating tip of 0.1‐mm radius of curvature penetrated into paraffin wax in two successive stages—fast and slow. After initial penetration due to loading, the sound was switched on and the penetration rate changed abruptly from zero to 500 μ/sec. After 40 msec, the fast rate then changed to a slower one that gradually decreased to zero. Somewhat similar results were obtained for other solids such as agar gel, tin, lead, and plastic. Sonic penetration may arise from high shearing stresses near the tip, which lead to plastic flow. Currently these stresses are being studied in plastics by means of stress birefringence. Findings may increase understanding of effects of ultrasound on solid like media in general. [Research supported in part by Hood Interdisciplinary Fellowship.]

ISSN:0001-4966    

DOI:10.1121/1.1977492

出版商:Acoustical Society of America    

年代:1971

数据来源: AIP

摘要:

Vehicular traffic noise continues to be a major source of complaint, especially near high‐speed highways. Of all vehicles on the road today, many consider trucks to be the major offender. Engine operating noise, exhaust noise, brake noise, and tire noise each contribute to the over‐all noise level produced by trucks; however, at speeds of 50 mph and greater, which is quite prevalent on today's interstate highways, the noise from tires predominates, provided the truck has a reasonably good exhaust muffler and is in a good state of repair. The lack of an information base of tire noise data available in the public domain has hampered the development of standardized tire‐noise testing procedures and the establishment of highway noise reduction criteria, standards, and regulations. This investigation will provide an inventory of data on representative tires and will develop an appropriate and accurate measurement methodology. The test parameters of this investigation include: tread design, tread depth, type of pavement, speed, and loading. A discussion is presented of measurement and analysis techniques leading to the attainment of (1) peakA‐weighted sound levels, (2)13‐octspectrum data, and (3) directionality information shown as equal sound‐level contour plots for various truck tires.

ISSN:0001-4966    

DOI:10.1121/1.1977498

出版商:Acoustical Society of America    

年代:1971

数据来源: AIP

摘要:

Machinery and appliances are usually enclosed by a panel structure, one of whose functions is to reduce the sound power radiated by the equipment. The theory employed in building acoustics for sound transmission through walls and partitions is not generally adequate for these situations since the sound field inside the enclosure is rarely diffuse, owing to the geometry of the machinery or appliance with respect to the enclosing surfaces. This paper outlines a theoretical approach that does not assume a diffuse sound field within the enclosure. Rather, it models the close dynamical coupling between the enclosure panels and the vibrating machinery inside. Also recognized is that the sound power radiated by the machine is affected by the enclosure, a fact that can be generally ignored in building acoustics. The theory shows the effect on acoustical performance of structural damping of the panels, acoustical absorption inside the enclosure, and the space between the machinery and the enclosure.

ISSN:0001-4966    

DOI:10.1121/1.1977505

出版商:Acoustical Society of America    

年代:1971

数据来源: AIP

摘要:

Experiments have been performed to investigate the simulation of cavitation bubbles by the implosion of thin‐walled glass spheres underwater. The bulbs were evacuated suitably gas‐filled if necessary, sealed, and then fractured underwater. High‐speed photography was used to study the implosion process and the radiated pressure waves were sensed by a hydrophone. The photographs revealed that the mode of collapse depended on the nature of the fracturing process. With a small impact, the glass is only punctured locally, and the glass appears to be fractured effectively by the inrushing water, giving rise to an unsymmetrical breaking. A more symmetrical fracture is obtained by using an externally generated shock wave. The possible buffering effect on the implosion of containing gas at appreciable pressure in a glass bulb was also investigated, together with the substitution of the surrounding water with a more viscous liquid. This type of experiment permits the variation and control of a number of parameters in the implosion phenomenon.

ISSN:0001-4966    

DOI:10.1121/1.1977509

出版商:Acoustical Society of America    

年代:1971

数据来源: AIP

摘要:

The classical statistical models for phonetic text view phonemes as states of a Markov chain of some low order. There is a model‐building method which produces models that lie between these classical models and the combinatorial model of O'Connor and Trimm. To produce the new models, one hypothesizes the existence of an underlying Markov chain of low order with some small numbernof states, and assumes that the probability distribution of phonetic symbols is a function of the state of the underlying chain. Givenn, the model‐building method involves a computer hill‐climbing procedure on a corpus of phonetic text to find the maximum‐likelihood model for that text and thatn. This paper describes the method and the results of a large number of ascents on a certain corpus of text. The models developed show strong linguistic features; for example, for text segmented only by silences between phoneme strings, the vowels, consonants, and boundary symbols are statistically identified with 15% errors. With text segmented into syllables, identification is essentially perfect.

ISSN:0001-4966    

DOI:10.1121/1.1977512

出版商:Acoustical Society of America    

年代:1971

数据来源: AIP