计算机辅助钢琴调律方法研究

发布时间：2018-06-05 12:01

本文选题：钢琴 + 调律　；参考：《辽宁大学》2017年硕士论文

【摘要】：现在,几乎所有领域都需要计算机技术的支持,音乐因运用计算机技术也取得很多成就。近年来计算机普遍运用于乐器声音调整。然而所有研究成果和软件都是支持音高调律的。一些钢琴生产家存在检查乐器的声音是否标准、悦耳,响度是否够大、够清晰、持续时间是否够长等需求,从而提出对钢琴结构、物料选择的调整方法,最终使钢琴声音最好听、最标准。然而现在的软件不能满足这些要求,所以钢琴声音特征调律的研究(如音色、音强、音长等)对于生产厂家具有重要意义,也能帮助普通用户检测钢琴声音是否标准。首先本文介绍了计算机音乐的发展历史。在部分上介绍了国内外钢琴调律研究和软件开发状况,以及具体乐器调律以及钢琴调律的工作原理。在主体内容,首先说到了音高调律,这部分介绍了基本的声波知识,接着说明频率、间距、人类感知、音高和频率的关系以及基本频率。进一步介绍了音高以及一些生理和物理特征对音高的影响。鉴于目前关于音高提取的研究比较完善,且现在也有很多比较好的钢琴调律软件,所以本文只探讨了现有提取音高的算法,并且通过实验选择最适合实现钢琴音高调律的算法。其次是对音色调律辅助的研究。这部分主要介绍音色的定义、特征以及音色特征与频率的关系。通过相关研究,让读者深入了解谱图。论文提出了两个支持音色调律的方法,分别是音色识别和音色比较,同时探讨了这两种方法的优缺点。音色识别方法使用音色特征样本来识别要调律的声音,关于此方法本文介绍了一些音色识别的算法。关于音色比较,本文提出了新方法,即运用模糊数学贴近度来比较两个音色频谱图,可以比较同一段时间内的两个频谱图的频带。运用模糊数学贴近度方法来比较音色谱图我们能够获得正确、详细的结果,且音色不受样本数量限制。最后是支持音强、音长的调律方法。本节从介绍音强、音长及其相关理论入手,并研究了计算音强与音长的方法。为了比较两个钢琴声音响度,我们不能直接用音强测量值来比较而是应用在过程中音强变化度来比较的,因为音强受空间距离和环境的影响很大。这部分本文提出用模糊数学贴近度来比较两个声音音强变化过程,根据5个模糊贴近度算法,对声音音强变化形态的6个案例进行了实验,找出最适用于音强比较的方法是最大最小贴近度。关于音长,本文就简单比较了两个声音的播放时间。
[Abstract]:Nowadays, almost all fields need the support of computer technology, and music has made a lot of achievements in the use of computer technology. In recent years, computers have been widely used in the sound adjustment of musical instruments. However, all research results and software support pitch. Some piano producers have the need to check whether the sound of the instrument is standard, pleasing to the ear, loud enough, clear enough, and long enough to last, so as to put forward a method of adjusting the structure of the piano and the choice of materials. Finally make the piano sound the best listening, the most standard. However, the current software can not meet these requirements, so the research of piano sound characteristic tuning (such as tone color, tone intensity, tone length and so on) is of great significance to manufacturers and can help ordinary users to detect whether piano sound is standard or not. First of all, this paper introduces the history of computer music. In part, it introduces the research and software development of piano tuning at home and abroad, as well as the working principle of musical instrument and piano modulation. In the main content, the first part is the pitch rhythm, which introduces the basic knowledge of sound waves, and then explains the frequency, distance, human perception, the relationship between pitch and frequency, and the basic frequency. The effects of pitch and some physiological and physical characteristics on pitch are further introduced. In view of the fact that the research on pitch extraction is relatively perfect, and there are many good piano tuning software, this paper only discusses the existing pitch extraction algorithms, and selects the most suitable algorithm for the piano pitch regulation through experiments. The second is the research of timbre and rhythm. This part mainly introduces the definition of timbre, characteristics and the relationship between timbre and frequency. Through the related research, let the reader deeply understand the spectrum chart. In this paper, we propose two methods to support timbre modulation, namely, timbre recognition and timbre comparison. At the same time, we discuss the advantages and disadvantages of these two methods. The timbre recognition method uses timbre feature samples to recognize the sound to be adjusted. This paper introduces some algorithms of timbre recognition. On the comparison of timbre, a new method is proposed in this paper, which is to compare two timbre spectrum maps by using fuzzy mathematical closeness, and to compare the frequency bands of two spectrum charts in the same period of time. Using the method of fuzzy mathematical closeness to compare the timbre spectrum we can obtain correct and detailed results and the timbre is not limited by the number of samples. Finally, it is to support the tone strong, the tone length of the tone method. This section begins with the introduction of tone intensity, pitch length and its related theories, and studies the method of calculating sound intensity and pitch length. In order to compare the sound acoustics of two pianos, we can't compare the sound intensity directly with the sound intensity measurement value, but apply the change degree of the sound intensity in the process, because the sound intensity is greatly affected by the space distance and the environment. In this part, we propose to compare the two sound intensity variation processes with fuzzy mathematical closeness. According to five fuzzy closeness degree algorithms, 6 cases of sound intensity change patterns are tested. The most suitable method for sound intensity comparison is maximum and minimum closeness. With regard to sound length, this paper briefly compares the playing time of two sounds.
【学位授予单位】：辽宁大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：J624.1;TP391.7

【参考文献】