[in Japanese]

View full abstract

We think that the evaluation of room acoustics should be done by a physical measure which relates also to the subjective evaluations. In this meaning, it is necessary to find the new physical measure containing the characteristics of hearing. From this view point we can enumerate Meyer's "definition", the ratio of signal to noise defined by Lochner and Burger and the ratio of time-sound pressure defined by us as new physical measures. It was reported that these measures correlate to the results of subjective evaluation (score of articulation and scale of tonal evaluation) more than to the reverberation time. The results of previous experiments on the tonal evaluation on which the selection of 50msec in the ratio of time-sound pressure is based have some possibility of further consideration in the following point. That is to say, though the psychological scales with the time delay of the first reflecting sound were measured in the experiments, it is considered that the psychological scales are also effected by the reverberant sounds because these vary with the time delay more or less. Lately we made experiments on the subjective tonal evaluation as to the time delay of the first reflecting sound and the level of reflecting sounds by the help of Scheffe's method of paired comparison in the following two cases. (1) Only the delay sound signal considered as the first reflecting sound is added to the direct signal sound. (2) The reflecting sounds, that consist of the first reflecting sound and the invariant reverberant sounds following it, are added to the direct sound signal. The artificial sound field synthesized in an anechoic room (Fig. 1) was used in the experiments on the tonal evaluation. We used the sound source signal recorded at a certain position in the hall as that of the second case. The histograms with no dashed lines in Fig. 7, 8 and 10 represent the results we have already reported (J. A. S. J. Vol. 19, No. 5 1963). Fig. 7 shows the psychological scales vs. the time delay in case that the speech sound is used as the source signal in the second case. Fig. 8 and Fig. 9 show the psychological scales vs. the time delay in case that the orchestra sound is used as the source signal in the second case. Fig. 9 shows the psychological scale of "splendid and vast" vs. the time delay in case that the hearing level is varied. Figs. 10 and 11 show the psychological scales vs. the time delay in case that the orchestra sound is used as the source signal in the first case. Fig. 11 shows the psychological scale of "splendid and vast" vs. the time delay in case that the hearing level is varied. Fig. 12 shows the psychological scales vs. the level of reflecting sounds in case that the time delay of the first reflecting sound signal is fixed at 50msec. The source signal is the same with the case of Fig. 7. Fig. 13 shows the psychological scale of "splendid and vast" vs. the level of reflecting sounds in case that the time delay of the first reflecting sound signal is fixed at 20msec and 50msec. Fig. 15 shows the psychological scales taken in the hall with R. T of 1. 1sec under the acoustical conditions represented in Table 1. The signal sound corresponding to the reflecting sounds in case of the artificial sound field goes through the reverberator and reproduced from the distributed loudspeakers in the hall shown in Fig. 14. The other orchestra sound as the source signal is provided. From the results on the time delay of the first reflecting sound, it may be said as following. The psychological scales of esthetic factor ("pretty") do not present a definite tendency to the delay time, but the psychological scales of the spatial and dynamic factor ("splendid and vast" and "resounding") show a definite tendency that varies at 40msec or 50msec. This tendency for the time delay is contrary to the case of

(View PDF for the rest of the abstract.)

View full abstract

In almost all the kinds of Japanese traditional music, the pitch intonation has been handed down by aural transmission of melodies. There is no standard pitch intonation. such as the twelve equal temperament of the western music, even today. And in this way various Japanese musics have been able to preserve their own characteristic pitch intonations. This paper is concerned with the pitch intonation of three kinds of Shamisen music. namely Naga-uta, Ji-uta, and Gidayu, studied by the experiments of measuring pitch of the practically demonstrated pieces. Measured samples are chosen out of wellknown pieces. They are: 1) Naga-uta about one hundred and eighty tones from a fragment of "Matsu-no-Midori", 2) Ji-uta about one hundred and forty tones from "Rokudan", 3) Gidayu about three hundred and seventy tones of "Ade-Sugata-Onna-Maiginu". They were played by each five players of the first class. The results are as follows: The consonant intervals such as perfect fourth, perfect fifth and octave are closely similar to the equal temperament, permitting exceptionally 17 cents stretch of perfect fifth in Gidayu. This justness of consonant intervals forms certain frames of melody and gives us some feeling of stability of melodious movement. Intonation characteristics of each music reveals within those frames, namely in intervals of semitone and wholetone, as summarized in Table 4. Fig. 1, 2 and 3 illustrate pitch intonation of melodies which show the interval characteristics typically. For instance, in the second example of Fig. 1, the semitone between the third note fis^1 and the fourth note g^1 and another one between the sixth note g^1 and seventh note fis^1 are narrow. And the two minor thirds enclosed by the semitones are wide as much. As for Fig. 2, the same deviation is observed in the semitone between the fifth note f^1 and eighth note e^1. In this case, the compensating stretch occurs in the major third between the fourth note e^1 and the fifth note f^1. In the example 1) of Fig. 3, wideness of wholetone is illustrated (e^1-d^1-e^1). Example 2) shows the wideness of wholetone and the compensating compression of minor thirds at the latter half ascending movement (a^1-h^2). Example 3) illustrates a specially wide semitone between g^2 and fis^2. Those melodies were demonstrated to eight specialists (musicologists and players) by a piano which had been tuned in the average intonations of each Shamisen music. They reported that even such presentation as was different in tone color from Shamisen playing made an impression of respective characteristic feeling of Shamisen music. In Tables 1, 2 and 3, the interval characteristics of three Shamisen musics are shown.

View full abstract