Machine Learning of Jazz Grammars

Gillick, Jon; Tang, Kevin; Keller, Robert M.

doi:10.1162/comj_a_00006

Cited by 25 publications

(37 citation statements)

References 12 publications

(9 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Later, Cruz-Alcázar and Vidal-Ruiz (1998) implemented several methods of grammatical inference: analyze a corpus of pre-existing classical music compositions, represented with a suitable set of symbols, then inducing stochastic regular grammars (Markov chains) able to parse the compositions in the corpus, and finally applying these grammars to generate new compositions that are in a similar style to the compositions in the corpus. Gillick et al (2009) used a similar approach (also Markovian) to synthesize jazz solos, but with a more elaborated synthesis phase. Kitani and Koike (2010) provide another example of grammatical inference, in this case used for real-time improvised accompaniment.…”

Section: Grammars and Related Methodsmentioning

confidence: 99%

“…Thornton (2009) defined a set of grammar-like rules from an existing composition, inferring a hierarchy of Markov models to use statistical patterns of the analyzed composition at multiple levels. As already mentioned in Section 3.1, Cruz-Alcázar and Vidal-Ruiz (1998) and Gillick et al (2009) progressions, (Davismoon & Eccles, 2010) used simulated annealing to combine constraints and Markov processes, and Pachet et al (2011) proposed a framework to combine Markovian generation of music with rules (constraints) to produce better results.…”

Section: Markov Chains and Related Methodsmentioning

confidence: 99%

See 1 more Smart Citation

AI Methods in Algorithmic Composition: A Comprehensive Survey

Fernández

Vico

2013

jair

209

113

View full text Add to dashboard Cite

Algorithmic composition is the partial or total automation of the process of music composition by using computers. Since the 1950s, different computational techniques related to Artificial Intelligence have been used for algorithmic composition, including grammatical representations, probabilistic methods, neural networks, symbolic rule-based systems, constraint programming and evolutionary algorithms. This survey aims to be a comprehensive account of research on algorithmic composition, presenting a thorough view of the field for researchers in Artificial Intelligence.

show abstract

Section: Grammars and Related Methodsmentioning

confidence: 99%

Section: Markov Chains and Related Methodsmentioning

confidence: 99%

AI Methods in Algorithmic Composition: A Comprehensive Survey

Fernández

Vico

2013

jair

209

113

View full text Add to dashboard Cite

show abstract

“…In the second example (see Figure 2), however, Parker clearly outlines both the E minor and the A7b9 chords in his solo line by arpeggiating the E minor chord and by emphasizing the flat 9 (Bb) of the A7b9 ehord. Often computer algorithms for improvisation are based on strict vertical relationships between the improvised line and the underlying chords so the line clearly reflects each chord (Band-in-a-Box, 2013; Gillick et al, 2010;Johnson-Laird, 2002;Ramalho et al, 1999). In two of these models, grammars based directly on the underlying chord progression are used to create improvised material, thereby possibly overemphasizing vertical elements in improvisational thinking (Gillick et al, 2010;Johnson-Laird, 2002).…”

Section: Dm"mentioning

confidence: 99%

“…First, using a corpus of improvised jazz solos, we compared the output from our algorithm with an improvisation created by an algorithm based on grammar and contour (Gillick et al, 2010;Keller, 2012). We found that our software creates improvisations whose embedded structure of patterns resembles that of human improvisers to a larger extent than the competing software.…”

Section: Jjj^^^^jjjmentioning

confidence: 99%

Testing cognitive theories by creating a pattern-based probabilistic algorithm for melody and rhythm in jazz improvisation.

Norgaard¹,

Spencer²,

Montiel³

2013

Psychomusicology: Music, Mind, and Brain

View full text Add to dashboard Cite

Building on previous work, which suggests that jazz improvisers insert patterns stored in procedural memory, a probabiMstIc model based on patterns from a corpus of Charlie Parker solos was developed and implemented. In previous analysis, patterns were detected in the corpus in significant proportions; however, the results of a parallel control situation showed minimal patterns. The control improvisation was generated by software based on grammars and contours, coincident with the cognitive position that emphasizes learned rule-based procedures in improvisation, as opposed to stored patterns. The present pattern-based improvisations, using our model, have graphs that coincide significantly with the actual human improvisation. Though briefly described earlier (Norgaard, Montiel, & Spencer, 20t3), the current article expands the theoretical foundation and adds methods for evaluating our algorithm using interval distributions and alternate corpora. Specifically, we show that the algorithm is capable of generating improvisations in fiddle and classical styles, demonstrating that the pattern-based algorithm is style independent. Our model shows much promise both for future research in the cognitive underpinnings of musical improvisation as well as for the development of software based on a stylistically appropriate concatenation of actual patterns.Perfortnance of preexisting tnusic and mtjsical improvisation both involve learned movements. However, during mu.sical improvisation, the exact configuration of those movements is determined in the moment. How is this accomplished? What information is stored in the improviser's brain that enables this complex behavior? One theory posits memorized schémas form the basis for the improvised output (Pressing, 1988), while a competing theory emphasizes learned rules (Johnson-Laird, 2002). The current project further explores these questions through the implementation of a computer algorithm for improvisation based on the principle advocated by Pressing. We compare output from our model with the results of a jazz analysis study as well as with the output of a competing model that uses a rule-based algorithm to generate melodies in a jazz style. In addition, we show that our algorithm is capable of generating melodic output in other styles given a corpus in that style.Pressing's (1988) model of the cognitive processes underlying improvisation is still widely cited (

show abstract

“…Contextspecific music generation in games had been widely used by LucasArts in their games with the help of a system known as iMUSE [LM94]. There have also been some computer-based composition software in the past like Band-in-a-box [Ban11] and Impro-Visor [GTK10] that have been quite popular.…”

Section: Motivationmentioning

confidence: 99%

Automatic Phrase Continuation from Guitar and Bass Guitar Melodies

Cherla

Purwins

Marchini

2013

Computer Music Journal

View full text Add to dashboard Cite

A framework is proposed for generating musically similar and interesting variations of a given monophonic melody. In this work, the focus is on rock/pop guitar and bass-guitar melodies with the aim of extensions to other instruments and musical styles. The original audio melody in audio format is first segmented into its component notes using onset detection and pitch estimation. Clustering is performed on the pitches to obtain a symbolic representation of pitch sequences in it. The note onsets are aligned with the estimated meter of the melody for a time-homogeneous symbolization of the rhythm in terms of onsets/rests and the metrical locations of their occurrence. A joint representation based on the cross-product of these two individual representations is used to train the prediction framework -the variablelength Markov chain. It is hypothesized that such a model will rearrange, while maintaining certain coherence, the segments of the original melody. The musical quality of the generated melodies was evaluated through a questionnaire by a group of experts and received an overall positive response.

show abstract

Machine Learning of Jazz Grammars

Cited by 25 publications

References 12 publications

AI Methods in Algorithmic Composition: A Comprehensive Survey

AI Methods in Algorithmic Composition: A Comprehensive Survey

Testing cognitive theories by creating a pattern-based probabilistic algorithm for melody and rhythm in jazz improvisation.

Automatic Phrase Continuation from Guitar and Bass Guitar Melodies

Contact Info

Product

Resources

About