摘要:

In the case of retrieving a sampled speech signal using a nonideal low‐pass filter, some loss of quality is observed due to passing energy grouped as side bands about the sampling frequency. This paper describes a series of experiments designed to relate the filter characteristic to the speech quality. By subjective matching of signals using filters with variable out‐of‐band transmission, it was found that the transmission in the vicinity of the sampling rate was the most important correlate to signal distortion. The transmission in the frequency region adjacent to the base band was found to be relatively unimportant. An explanation in terms of the spectral characteristic of speech signals is offered. Presampling low‐pass filtering was also investigated and found to be much less important than accurate post sampling filtering.

ISSN:0001-4966    

DOI:10.1121/1.1936047

出版商:Acoustical Society of America    

年代:1959

数据来源: AIP

摘要:

Two series of formal listening tests were conducted with fricative consonants generated by a dynamic analog of the vocal tract. The consonants were generated using a fricative configuration consisting of a uniform tube with a single constriction and were presented in CV context, always with the vowel /ɑ/. The variables studied in the first test series were the place and degree of constriction and the position of noise excitation relative to the constriction. Listeners were required to make one of six fricative responses in forced and nonforced tests. Responses of /s,∫,f/ better than 95% were obtained for some stimuli. The second series of tests was made to determine how closely the temporal relations between articulatory change and buzz and noise excitation need to be specified. Stimuli were presented in pairs and subjects were asked to vote for the more “natural” stimulus in each pair. A supplementary identification test showed a sharp change of response from voiceless to voiced fricative as buzz excitation was initiated at progressively earlier times. [This work was supported in part by the U. S. Army (Signal Corps), the U. S. Air Force (Office of Scientific Research, Air Research and Development Command), and the U. S. Navy (Office of Naval Research) and in part by Air Force Cambridge Research Center Contract AF 19(604)‐2061.]

ISSN:0001-4966    

DOI:10.1121/1.1930226

出版商:Acoustical Society of America    

年代:1959

数据来源: AIP

摘要:

Unison strings of a concert grand piano were tuned to five “unison” conditions. The conditions were “zero‐beat” tuning and the upper string of three string unison groups tuned sharp and the lower string tuned flat by12, 1, 2, and 3 cents relative to the center string. Magnetic tape recordings were made of the piano tuned under these conditions. These recordings in the form of a paired comparison preference test were presented to musically trained and untrained subjects. The most preferred tuning conditions for three string unison groups are 1 and 2 cents maximum deviation among strings. Musically trained subjects prefer less deviation in tuning among unison strings than do untrained subjects. Close agreement was found between the subject's tuning preferences and the way artist tuners actually tune piano unison strings. The results of this investigation are interpreted as supporting a “set” or learning hypothesis concerning the origin of preferences for piano unison tuning.

ISSN:0001-4966    

DOI:10.1121/1.1930231

出版商:Acoustical Society of America    

年代:1959

数据来源: AIP

摘要:

While gathering data on the masking of one pure tone by another, marked pitch shifts in the signal tone were noted. This paper presents preliminary results of a systematic investigation of this phenomenon. Signals of five frequencies and two levels were used together with eight masker frequencies for each signal frequency. For a given condition signal level was constant, but the signal was effectively raised above masked threshold by lowering masker level. For three masker levels listeners adjusted a matching tone, heard in the contralateral ear, to the pitch and loudness of the masked tone. Stimuli were presented alternately to each ear. The results indicate that with no masking a just noticeable difference (JND) in frequency is much larger when the stimuli to be discriminated are presented to opposite ears rather than to the same ear. Further, this JND increases rapidly for frequencies above about 3200 cps. Under conditions of masking, the results are similar to those obtained using a narrow band noise masker. In general, the pitch of the signal is shifted away from the frequency region of the masker. The maximum shift occurs when signal and masker frequencies are nearly equal. Also, as signal threshold is approached the amount of pitch shift increases. Possible relations between the slope of the masking function and the magnitude of the pitch shift are discussed.

ISSN:0001-4966    

DOI:10.1121/1.1930236

出版商:Acoustical Society of America    

年代:1959

数据来源: AIP

摘要:

Storing data from acoustical experiments so that they may be recovered or played back for precise laboratory selection and analysis has always been a difficult problem. A method is described in which a continuous signal is sampled, quantized, coded, and recorded on magnetic tape in the form of straight binary digital numbers of 10‐bits plus sign. These numbers are then handled using digital techniques. The method uses a start‐stop recorder, sequential core memory, digital computer, computer memory scanner, and a plotter. The core memory performs the buffering of data from the recorder to the computer memory. In operation the playback digital recorder fills the core buffer with a block of 100 words at some convenient rate and stops. The block of data is then transferred electrically from the core buffer to the computer memory at the synchronized word rate of the computer. To aid in data selection, a digital‐to‐analog converter and an oscilloscope display are used to permit scanning the data visually while in either the core memory or the computer memory. An associated plotter is used to plot out the original data and the results of analysis. All operations of external equipment are under the control of special commands stored in the computer memory as subroutines along with arithmetic processing routines. This arrangement allows the handling and the arithmetic programs to be intermixed in any combination of data selection and processing that is desired. An example of data selection and arithmetic analysis is described.

ISSN:0001-4966    

DOI:10.1121/1.1930246

出版商:Acoustical Society of America    

年代:1959

数据来源: AIP

摘要:

Phoneticians have tried in various ways to investigate the maintenance of tonal distinctions in whispering, where there is no variation of a fundamental frequency. In the present experiment it was found with whispered speech that native speakers of Thai, a language with five phonemic tones, could identify words minimally distinguished by tone, though with somewhat less accuracy than in normal speech. To determine whether such concomitant features as helped the subjects to make systematic distinctions in the whispered speech were also present in normal speech, the spoken words were passed through the Vocoder with the fundamental frequency of the buzz kept constant. No discriminations were made! With hiss alone, however, the results were better than in the natural whisper. We conclude that the features were present in the normal speech but that in the presence of the buzz, listeners were set to hear pitch variations. Inspection of spectrograms suggests that tonal oppositions in Thai whispering lean on such concomitant features as changes in intensity, relative durations of vowels, and small variations in formant frequencies.

ISSN:0001-4966    

DOI:10.1121/1.1930248

出版商:Acoustical Society of America    

年代:1959

数据来源: AIP

摘要:

The rating method was used to obtain operating characteristics for sixty heterogeneous words. A single message was heard in noise, or seen briefly in a tachistoscope. It was repeated until it had been assigned to the highest accuracy category (“confirmed”) or sent a maximum of six trials. The comparisons showed that it matters little whether reception is by eye or by ear. Whether within a trial or over successive repetitions, accuracy of reception is a direct function of the confidence rating and is relatively independent of the intelligibility level. Neither do the accuracy of reception or the distributions of rating categories change markedly over trials. Although no direct test was made, it appears that accuracy of reception is not lessened by the task of rating. Both visual and auditory data are fitted reasonably well by predictions made from a simple stochastic model based on the assumptions that (1) intelligibility, (2) probability of a correct acceptance, and (3) probability of an incorrect acceptance remain constant over successive repetitions. The model fits the visual better than the auditory data, as might be expected, since conditions of reception are more homogeneous over trials for vision than for audition. [This research was supported by the Group Psychology Branch, Office of Naval Research under Contract No. Nonr 233 (58).]

ISSN:0001-4966    

DOI:10.1121/1.1930258

出版商:Acoustical Society of America    

年代:1959

数据来源: AIP

摘要:

A unique facility has been constructed for testing flight vehicle structures and electronic systems in the presence of high intensity sound. The siren sound source will produce pure tones or narrow bands of noise throughout the frequency range from 50 to 10 000 cps with controlled modulation from 0 to 50 cps. The maximum acoustic power output is 25 000 w to produce a sound pressure level of approximately 174 db in the 1 ft square progressive‐wave test section. This paper discusses the performance of the facility. (The design and construction of this facility was sponsored by the Electronic Technology Laboratory and the Aircraft Laboratory of Wright Air Development Center.)

ISSN:0001-4966    

DOI:10.1121/1.1930265

出版商:Acoustical Society of America    

年代:1959

数据来源: AIP

摘要:

The extension of digital techniques to the problem of processing the output of an array of point receiving elements in an acoustic field has given rise to a method of phasing called DIMUS (Digital Multi‐beam Steering). This digital technique incorporates the use of shift registers as delay line elements which results in a number of operational advantages. A description of the DIMUS technique is given. The effects of amplitude and time quantization imposed by the digital circuits are considered in the light of the relative signal processing gain of the array referred to an exact analog phasing network. The processing gain is computed as a function of sampling frequency for a number of types of signals. [This work represents one of the results of research carried out under Contract Nonr 2216 (10) with the Office of Naval Research.]

ISSN:0001-4966    

DOI:10.1121/1.1930268

出版商:Acoustical Society of America    

年代:1959

数据来源: AIP

摘要:

The well‐known SOFAR channel manifests itself in modified form for sound sources situated off axis. Sources are usually closer to the ocean surface than to the axis, and in such cases high‐intensity “convergence zones” may appear at spacings of some 20 to 35 mi depending on latitude and on velocity conditions. Formed usually by deep “refracted‐surface reflected” rays the caustic structure of a zone is known. These convergence zones are analogous to the zones of sound which alternate with zones of silence in abnormal sound propagation through the atmosphere. Recent publications giving the ray theory and specific details of the uw sound propagation have created a situation in which both much and little information is available. So that this situation may be relieved, a review of both experimental and theoretical efforts will be attempted. These efforts were carried out by both government and contracting laboratories. Accomplishments in recent years have shown that convergence zone propagation is common and predictable. Methods of fairly precise intensity computation exist but entail considerable effort. The existing computations, theory and experiments provide enough information so that convergence‐zone propagation in the first several hundred mile range interval can be adequately described. Merging and confusion of zones is to be expected at greater ranges.

ISSN:0001-4966    

DOI:10.1121/1.1930271

出版商:Acoustical Society of America    

年代:1959

数据来源: AIP