A frequency compression technique for the hearing impaired was studied. In order to elucidate the validity of the frequency compressed speech for the hearing impaired, intelligibility test was made with simulated hearing loss for normal hearing subjects. Results showed that (1) the intelligibility was improved by the frequency compression for vowels, but for consonants, (2) the optimum frequency compression ratio varied depending on the individual hearing loss characteristics as well as on the formant and the fundamental frequencies of the input speech, (3) lowering of the fundamental frequency resulted in higher intelligibility in the frequency compressed speech, and (4) the improvement in intelligibility was observed in the particular vowels.
A noise or vibration system must be usually considered as a non-linear system with a finite memory, if taking a total system including sound (vibration) source, propagation mechanism and observation mechanism into consideration. In this paper, a new practical method of statistically evaluating an output probability distribution of the above system is derived consistently in one-variate form by using positively the well-known Wiener's idea on a generalized analysis of a non-linear system. That is, first, the objective system is approximately separated into two functional parts, i. e. , a zero-memory type non-linear part and a finite memory type linear part. Then, the output probability distribution can be evaluated in an approximate but generalized form, based on their typical functions. The effectiveness of the proposed method is experimentally confirmed by applying it to an actual room acoustics.
It has been well-known in radar technology that the intensity of echoes scattered by a random medium can be evaluated quantitatively in term of a backscattering coefficient. In order to apply this concept to ultrasonic pulse echo techniques, the influence due to the aperture length of window must be compensated since in ultrasonics the broad band pulse is used so as to increase the resolving power of the system. Thus the measurement of backscattering coefficient must be dependent on the window length as well as the duration of ultrasonic pulse. In this paper the influence of these factors on the measurement of backscattering coefficient is analyzed under the assumptions of Gaussian function for the window, the pulse and the correlation function of tissue. And then the results are generalized to give a formula for in vivo measurement of backscattering coefficient for arbitrary window functions. Finally the availability of the formula is shown in the experiments where a sponge is used as the random medium.
The slow-tube waveguide, a new type of the acoustic slow waveguide radiator, is described. In this radiator, a thin-walled slow-tube waveguide is used as the basic element in an end-fire radiator. By introducing the equivalent source function within the slow-tube, the determination of the sound radiation field is conveniently carried out from the volume integral, not the surface integral. It is ascertained that Hansen-Woodyard's condition giving the optimum length of the slow-rod waveguide radiator is also valid for the slow-tube waveguide radiator. The effects of the wall thickness for long tube on the directional pattern are investigated theoretically and the superiority of the thin-walled slow-tube is confirmed. The poor directional pattern of the thick-walled slow-tube is related to the wave number ratio (k_z)/k differing substantially from (k_z)/k=1, where k_z is the wave number along the slow-tube and k the wave number of the surrounding medium. At the same time, it is clarified that the large diameter slow-tube leads to increase in directivity, especially by lowering the side lobe level.
Since 1982 we have been surveyed daily noise exposure and associated informations of residences in Nagoya city. Classification of measured noise patterns of residences has been made by using cluster analysis. The five typical groups of patterns are shown to exist and they are mainly related to road traffic conditions and land use nearby residence. Then in this paper noise ratings based on L_<eq> are also examined with respect to reaction of inhabitants. Scalings of reaction have been done by using Likert scale and the successive category method, on which one can quantitatively discuss the relation between L_<eq> and reaction of inhabitants to noise environment. The results show that as for L_<eq24> reactions of inhabitants are gentle below 55 dBA, moderate at 60 dBA, but severe over 65 dBA. The other noise ratings such as L_<dn>, L_<eq>M, L_<eq>D, L_<eq>E, L_<eq>N and L^^<(a)>__<eq1/6> are also considered.