Training Supports for the Japanese Classic Music by Audio Signal Processing.

在漫长的和平时期培养出来的一些复杂的日本艺术，将随着西方化和战后的混乱而逐渐消失，失去其身份。在独特的文化中，日本古典音乐面临着危险，很少有人欣赏它的价值。为了防止传统音乐的彻底消失，我们需要做出很大的努力。本研究试图通过对音频信号处理的分析，为日本古典音乐的培训和教育提供一些辅助方法。利用个人电脑修改计算机语音识别软件，分析了日本乐器的特点，并对演奏者的水平进行了分类：日本音乐与西方音乐有几个不同的特点，这些特点可以用谱图来表现。通过A/D转换器以10微秒的采样率连续采集到RAM板上的声音，在个人计算机上进行谱密度分析。比较了不同层次在shakuhati音上的能力，shakuhati音被认为很难演奏。研究人员还研究了日本古典音乐的电子键盘，并将其与用彩色谱图显示的真琴进行了比较。这些研究得出了一些结论，即日本古典音乐的专家和初学者之间的差异，在日本古典音乐中，声音更多地取决于个人演奏者，而不是西方同行，可以通过彩色谱图清楚地区分出来。音乐合成器也可以发出和shakuhati相似的声音。另一项研究是关于使用X-Y跟踪器获取嘴唇运动的嘴部信息的语音识别。这项研究可能有助于为shakuhati运动员提供支持系统，以获得颈部挥舞的位置，这是很难的，因为它需要三年的训练。最后，我希望有关方面给所有小学和/或初中教授日本音乐的机会。少

Some sophisticated Japanese arts, cultivated in a very long peace time, will fade away and lose its identity by westernization and the postwar disorders. Among the unique culture, the Japanese classic music have faced a danger, where very few people appreciate its value. Some great efforts are needed to prevent complete loss of our traditional music.This study has tried to provide some supporting methods for training and education of the Japanese classic music by analysis of audio signal processing. Modifying computer software for speech recognition using a personal computer, we analyzed characteristics of Japanese musical instruments together with the classifying players' levels : the Japanese music has several specific characters different from western music, which can be shown by spectrograms.Sounds continuously taken by an A/D converter with 10 micro second sampling rate into a RAM board are processed to make spectral density analysis in a personal computer. Differences among the p … More layers' abilities on the shakuhati sounds, which is said to be difficult to play, were compared. An electronic keyboard for Japanese classic music was also examined to make a comparison with a real shakuhati, that was shown by color spectrogram. These studies have lead to some conclusions that the differences between the experts and the beginners in Japanese classic music, where the sounds depends on the individual players more than the western counterparts, are clearly distinguished by the color spectrogram. And it was also shown that the music synthesizer can make similar sounds of the shakuhati.The other study was about the speech recognition with mouth information using an X-Y tracker which obtains lip movements. This study might help give supporting systems for the shakuhati players to get the positions for neck waving which is difficult as saying that it requires three year training. As a final remark, I would like the concerned to give chances for teaching Japanese music to all the elementary and/or the junior high schools. Less