Skip to main content

2014 | Buch

Computational Musicology in Hindustani Music

insite
SUCHEN

Über dieses Buch

The book opens with a short introduction to Indian music, in particular classical Hindustani music, followed by a chapter on the role of statistics in computational musicology. The authors then show how to analyze musical structure using Rubato, the music software package for statistical analysis, in particular addressing modeling, melodic similarity and lengths, and entropy analysis; they then show how to analyze musical performance. Finally, they explain how the concept of seminatural composition can help a music composer to obtain the opening line of a raga-based song using Monte Carlo simulation.

The book will be of interest to musicians and musicologists, particularly those engaged with Indian music.

Inhaltsverzeichnis

Frontmatter
Chapter 1. An Introduction to Indian Classical Music
Abstract
The origins of Indian classical music lie in the cultural and spiritual values of India and go back to the Vedic Age (Sam Veda). Even in those times, music was handed down orally from the guru (teacher) to the shishya (disciple). The art was called sangeet and included vocal music, instrumental music, and dance. The great sages who dwelt in ashramas (hermitages) imparted instruction to their students who lived with them on the premises. The art of music was regarded as holy and heavenly. It not only gave aesthetic pleasure but also induced a joyful religious discipline. Devotional music was intended to take man towards God and give him an inner happiness and self-realization. Subsequently this art branched off into three separate streams: vocal music (geet), instrumental music (vadya), and dancing (nritya).
Soubhik Chakraborty, Guerino Mazzola, Swarima Tewari, Moujhuri Patra
Chapter 2. The Role of Statistics in Computational Musicology
Abstract
Music analysis, broadly speaking, can be divided into commonality analysis (“what is common?”) and diversity analysis (“what is special?”). It is worth pointing out here that a statistician is essentially a commonality expert in the sense that the philosophy of statistics is to summarize and average and make inferences which are true on the whole and which describe a process rather than an individual entity. Fortunately, there are issues in music where this traditional mindset of the statistician finds a support. For example, a collection of recordings of the same artist if analyzed statistically will definitely reflect certain common features having to do with the style of the artist. But the statistician must realize that every single music piece will have something special to offer. Fortunately, again, there are issues even in statistics where the statistician does take an individual observation seriously—as in the case of an outlier or influential observation, for example. There is a whole literature in statistics to deal with outliers. When an outlier comes, the traditional philosophy of summarizing and averaging is brushed aside. The statistician goes after this individual influential observation exploring how it came and what it signifies. The case of outliers is an exception in statistics. It is the very grammar in music as music is a work of art! If the statistician can use his experience and mindset of handling outliers (regarding every musical piece as a musical outlier) along with his commonality expertise, he can be a very effective music analyst. Similar point of view has been expressed by Nettheim who has also provided a good bibliography of statistical applications in musicology (Nettheim 1997). For a sound statistical treatment of musical data, see Beran and Mazzola (1999). Additionally, we acknowledge the contributions from Meyer (1989), Snyder (1990), Winkler (1971), Wilson (1982), Todd and Loy (1991), and Morehen (1981). It is an irony that computational musicology in Indian classical music is still lagging behind the progress in Western classical counterpart, although we do appreciate the efforts of Castellano et al. (1984), Chordia and Rae (2007), and Sinha (2008) among others. We hope this book will provide some food for thought in that direction. Statistics is a useful tool both for analyzing a musical structure and quantitative assessment of a musical performance. The former helps in revealing features of a musical piece in general, while the latter brings the style of the artist into consideration as well in performing the musical piece.
Soubhik Chakraborty, Guerino Mazzola, Swarima Tewari, Moujhuri Patra
Chapter 3. Introduction to RUBATO: The Music Software for Statistical Analysis
Abstract
RUBATO® is a metamachine designed for representation, analysis, and performance of music. It was developed on the NEXTSTEP environment during two SNSF grants from 1992 to 1996 by the author and Oliver Zahorka (SNSF Research Reports. Universität Zürich, 1993–1995; Proceedings of the ICMC 94. ICMA, 1994), Mazzola et al. (Computing in Musicology. CCARH, 1995b; The RUBATO homepage. Univ. Zürich, 1996), Zahorka (Animato 97(3): 9–10, 1997a; Symposionsband Klangart ’95. Schott, 1997b). From 1998 to 2001, the software was ported to Mac OS X by Jörg Garbers in a grant of the Volkswagen Foundation. RUBATO®’s architecture is that of a frame application which admits loading of an arbitrary number of modules at run-time. Such a module is called RUBETTE®. There are very different types of Rubettes. On the one hand, they may be designed for primavista, compositional, analytical, performance stemma, or logical and geometric predication tasks. On the other, they are designed for subsidiary tasks, such as filtering from and to databases, information representation, and navigation tasks, or else for more specific subtasks for larger “macro” Rubettes. A RUBETTE® of the subtask type is coined OPERATOR and implements, for example, what we have called performance operators in section [Mazzola et al. (The Topos of Music. Birkhäuser, 2002, 44.7)]. The RUBATO® concept also includes distributed operability among different peers. This software is conceived as a musicological research platform and not a hard-coded device; we describe this approach. Concluding this chapter, we discuss the relation between frame and modules.
Soubhik Chakraborty, Guerino Mazzola, Swarima Tewari, Moujhuri Patra
Chapter 4. Modeling the Structure of Raga Bhimpalashree: A Statistical Approach
Abstract
Music, according to Swami Vivekananda, is the highest form of art and also the highest form of worship (provided you understand it!). Understanding music, both esthetically and scientifically, becomes important. This is especially true for classical music, be it Indian or Western, since each is a discipline in its own right. While the former stresses on the emotional richness of the raga as expressed through melody and rhythm, the latter is technically stronger as, in addition to melody and rhythm, the focus is also on harmony and counterpoint. A raga is a melodic structure with fixed notes and a set of rules characterizing a particular mood conveyed by performance. The present chapter gives a statistical structure analysis of raga Bhimpalashree.
Soubhik Chakraborty, Guerino Mazzola, Swarima Tewari, Moujhuri Patra
Chapter 5. Analysis of Lengths and Similarity of Melodies in Raga Bhimpalashree
Abstract
In monophonic melody analysis, significance of a melody is measured by multiplying its length with the number of times it occurs. An analysis of melody lengths begins a new line of investigation (for a more comprehensive account, see Chakraborty et al. 2011–2012; this is possible in the first work on melody lengths). The strength of our results stems from the fact that a structure analysis discovers several interesting facts of the raga without restricting to any specific artist. The sequence of the raga notes, taken from a standard text, is given in Appendix of Chap. 4. A raga is a melodic structure with fixed notes and a set of rules characterizing a certain mood conveyed by performance. Two melodies, of equal length, are similar if the correlation coefficient of their shapes is significant. These concepts are detailed in the next section.
Soubhik Chakraborty, Guerino Mazzola, Swarima Tewari, Moujhuri Patra
Chapter 6. Raga Analysis Using Entropy
Abstract
If P(E) is the probability of an event, the information content of the event E is defined as I(E) = −log2(P(E)). Events with lower probability will signal higher information content when they occur. The probability of a raga note, and hence its information content, depends on the raga concerned. The important raga notes will obviously be having higher probabilities. On the other hand, a weak note in a raga cannot be thrown away either for it would be carrying more surprise! The strength of entropy analysis lies here (entropy is the mean information content of a random variable).
Soubhik Chakraborty, Guerino Mazzola, Swarima Tewari, Moujhuri Patra
Chapter 7. Modeling Musical Performance Data with Statistics
Abstract
Today, the relationship between music and mathematics is a common factor. In the last two or three decades, the advances in mathematics, computer science, psychology, semiotics, and related fields, together with technological progress (in particular computer technology), lead to a revival of quantitative thinking in music [see, e.g., Archibald (1972), Babbitt (1961), Balzano (1980), Lewin (1987), Lendvai (1993), Forte (1964, 1973), Morris (1987, 1995), Johnson and Wichern (2002), Leyton (2001), Andreatta (1997), Solomon (1973), Beran and Mazzola (1999), Meyer (1989)]. Musical events can be expressed as a specific ordered temporal sequence, and time series analysis is the observations indexed by an ordered variable (usually time). It is therefore not surprising that time series analysis is important for analyzing musical data as it is always be the function of time. Music is an organized sound. But the equation of these sounds does not produce the formula of how and why sounds are connected. Statistics is a subject which can connect theoretical concept with observable phenomenon and statistical tools that can used to find and analyzing the structure to build a model. But applications of statistical methods in Indian musicology and performance research are very rare. There were some researches that had been done on Western musicology and mostly consist of simple applications of standard statistical tools. Due to the complex nature of music, statistics is likely to play an important role where the random variables are the musical notes which are function of time.
Soubhik Chakraborty, Guerino Mazzola, Swarima Tewari, Moujhuri Patra
Chapter 8. A Statistical Comparison of Bhairav (a Morning Raga) and Bihag (a Night Raga)
Abstract
The present chapter gives an interesting statistical comparison of two ragas, namely, Bhairav (morning raga) and Bihag (night raga). Metric and melodic properties of notes are analyzed using RUBATO along with distinct transitory and similar looking non-transitory pitch movements (but possibly embedding distinct emotions!) between the notes for both the ragas. According to Strawn (1985), “a transition includes the ending part of the decay or release one note, the beginning and possibly all of the attack of the next note and whatever connects the two notes.” Hence in addition to the study on modeling a performance, a count for distinct transitory and similar looking non-transitory frequency movements (but possibly embedding distinct emotions!) between the notes was also taken. In the characterization of ragas in Indian music, not only the notes and note sequences but how they are rendered are important. There is a concept of alankar in Indian music meaning ornament (of course in a musical sense!). The shastras have categorized alankars into Varnalankar and Shabdalankar (http://​www.​itcsra.​org/​alankar/​alankar.​html). The varnas include sthayi (stay on a note), arohi (ascent or upward movement), awarohi (descent or downward movement) and sanchari (mixture of upward and downward movement). This classification of alankars relate not only to the structural aspect of the raga but also to the raga performance. A non-transitory movement would depict stay on a note for short or long duration. In the graph it looks close to a horizontal line with some tremor (the tremor is because pitch is never steady).
Soubhik Chakraborty, Guerino Mazzola, Swarima Tewari, Moujhuri Patra
Chapter 9. Seminatural Composition
Abstract
Music composition can be natural, artificial, or seminatural. The first category refers to a composition where a human being decides both what to play (or sing) as well as how to. The second refers to a composition where a mechanical device such as a computer is trained to accomplish both the what and the how part. The third category is of interest here in which the computer will decide the what part, while a human being will take up the how part. Consider a music composer interested in composing a raga-based song. He is looking for the starting line or a clue for the next line. Can computer help? We answer this question through seminatural composition giving an example in raga Bhimpalashree. Seminatural composition was introduced in Chakraborty et al. (2009) and is discussed more formally in Chakraborty et al. (2011) recently. However, for the sake of completeness, we are providing below the algorithm seminatural composition algorithm (SNCA) after which the illustrative example will follow. For an extensive literature on algorithmic composition, the reader is referred to Nierhaus (2008).
Soubhik Chakraborty, Guerino Mazzola, Swarima Tewari, Moujhuri Patra
Chapter 10. Concluding Remarks
Abstract
What is primarily common between music and statistics? I think it is the fact that a musician is imagining and creating musical patterns during a composition. Statistics, on the other hand, is the science of exploring and studying patterns in numerical data. Musical data are certainly numerical in character as they pertain to pitch, onset and departure of notes, loudness, timbral characteristics, etc. All these can be subjected to a careful statistical analysis. I have also worked, and am still working, with a team of doctors and another statistician studying the therapeutic impact of Hindustani ragas on patients with brain injury, and I can assure you that it is very difficult, if not impossible, to establish the aforesaid impact without a sound statistical analysis, even if we all know music can heal.
Soubhik Chakraborty, Guerino Mazzola, Swarima Tewari, Moujhuri Patra
Metadaten
Titel
Computational Musicology in Hindustani Music
verfasst von
Soubhik Chakraborty
Guerino Mazzola
Swarima Tewari
Moujhuri Patra
Copyright-Jahr
2014
Electronic ISBN
978-3-319-11472-9
Print ISBN
978-3-319-11471-2
DOI
https://doi.org/10.1007/978-3-319-11472-9