This paper describes the use of comb type quefrency window in the cepstrum analysis. A good result is obtained with a Hanning lifter in case of low pitch frequency. But in case of high pitch frequency, it is sometimes impossible to separate two formants. The comb lifter is presented to overcome this difficulty; that is, the lifter length is fixed so as to separate the closest formants, and the peaks in the cepstrum due to the periodicity of speech are suppressed using the comb lifter. It is necessary for the utilization of the comb lifter to estimate exact pitch period. Since it is hard to estimate an exact pitch period, the pseudo-comb lifter which does not need the exact pitch period is proposed. For the case of the high pitch frequency, the comb lifter is effective and for the case that the pitch period is undetectable, the proposed pseudo-comb lifter is effective.

View full abstract

The propagation of the finite amplitude wave through the pipe system which is made of the infinite long pipe with two discontinuity cross sections is dealt with in this paper. At first, the measurement of the nonlinear response of the system is made by the use of the impulsive incident wave similar to a period of sinusoidal waveform, and then the simulation of the transmitted wave through the system is performed under some assumptions to compare the experimental results. In the simulation, the linear theory is used in the operation of the reflection and the transmission at the boundary, and the nonlinear theory is used in the operation of the propagation through the pipe. The simulated results agreed with the observed results well. On the other hand, considerable disagreement between simulated results and the linear theory are found around a certain frequency given by dip in the spectrum, and so it may be concluded that the nonlinear response of the system is determined by the waveform distortion due to the finite amplitude effect with propagation through the pipe, and therefore the response depend upon not only the amplitude but also the pulse-length of the incident wave.

View full abstract

An analytical approach is presented to examine the mechanisms of the transient sound fadiation from a clamped circular plate for impulsive sound wave. After a bfief review on the derivation of the formulae by which the fairly accurate transient waveform radiated is obtained in the time domain, this mechanism is examined in detail in terms of the waveform radiated when the plate is excited by the plane triangular sound pulse of normal incidence. The felationship between the mechanisms of the sound radiation and the mechanical properties of the plate is given and it is noted that one can obtain an arbitrary waveform when the mechanical constants of the plate are controlled.

View full abstract

A new method is presented for the measurement of normal incident sound absorption characteristics of materials using an acoustic tube. As the measuring principle, by measuring the signals of sound pressure at two points in the acoustic tube using two microphones, and by processing these signals either in time domain or in frequency domain, the incident wave and the reflected wave are equivalently separated. For the sound source, any of pure tone, random noise or impulse can be used in this measuring method. As a result of the experimental study by making measurements on two types of sound absorbing materials, it is ascertained that almost the same results can be obtained by this method as those measured by the ordinary standing wave method.

View full abstract

Based on the Mathieu function method, the symmetry property of the ultrasonic lightdiffraction spectrum with respect to the angle of incidence around the Bragg incidence angle is investigated. The correspondence rule of the diffraction spectrum is extended for the non-Bragg incidence case and it is shown that the reciprocity theorem of ultrasonic light-diffraction phenomenon is also valid for the non-Bragg incidence, so that the theorem is valid for any angle of incidence, including Bragg incidence and non-Bragg incidence.

View full abstract