Functional Harmonic Analysis Using Probabilistic Models

Raphael, Christopher; Stoddard, Joshua

doi:10.1162/0148926041790676

Cited by 75 publications

(53 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These procedures were implemented by Sleator and Temperley as part of the Melisma Music Analyzer [38], which is mainly based on preference rules and described in more detail below. In [39], the authors presented a hidden Markov model that uses symbolic MIDI data as input and produces a harmonic analysis of a musical piece including key and roman numerals labeling. Reference [40] describes a chord labeling system for MIDI guitar sequences that is based on the symbolic chord labeler proposed by [41].…”

Section: Automatic Chord Labelingmentioning

confidence: 99%

Towards Cross-Version Harmonic Analysis of Music

Ewert

Müller

Konz

et al. 2012

IEEE Trans. Multimedia

View full text Add to dashboard Cite

Abstract-For a given piece of music, there often exist multiple versions belonging to the symbolic (e.g., MIDI representations), acoustic (audio recordings), or visual (sheet music) domain. Each type of information allows for applying specialized, domain-specific approaches to music analysis tasks. In this paper, we formulate the idea of a cross-version analysis for comparing and/or combining analysis results from different representations. As an example, we realize this idea in the context of harmonic analysis to automatically evaluate MIDI-based chord labeling procedures using annotations given for corresponding audio recordings. To this end, one needs reliable synchronization procedures that automatically establish the musical relationship between the multiple versions of a given piece. This becomes a hard problem when there are significant local deviations in these versions. We introduce a novel late-fusion approach that combines different alignment procedures in order to identify reliable parts in synchronization results. Then, the cross-version comparison of the various chord labeling results is performed only on the basis of the reliable parts. Finally, we show how inconsistencies in these results across the different versions allow for a quantitative and qualitative evaluation, which not only indicates limitations of the employed chord labeling strategies but also deepens the understanding of the underlying music material.

show abstract

Section: Automatic Chord Labelingmentioning

confidence: 99%

Towards Cross-Version Harmonic Analysis of Music

Ewert

Müller

Konz

et al. 2012

IEEE Trans. Multimedia

View full text Add to dashboard Cite

show abstract

“…Raphael & Stoddard [24] proposed a machine learning approach based on a Hidden Markov Model that computes Roman numeral analysis (that is, the higher level, functional analysis mentioned above). A main feature of their system is that the generative model can be trained using unlabeled data, thus determining its applicability also to unsupervised problems.…”

Section: Related Workmentioning

confidence: 99%

“…ChordDistance m 1 i m 2 features identify roughly three kinds of transitions: 1) those providing strong positive evidence in favor of a given label y t ; 2) those used as refinement criteria; 3) those used to forbid the transitions that have associated large negative numbers. All transitions involved in the cadence have been identified among the most relevant horizontal features (see features number 30,22,34,25,26,32,23,24 in Table 2) and Figure 6.…”

Section: Musical Interpretation Of Acquired Classifiersmentioning

confidence: 99%

BREVE: An HMPerceptron-Based Chord Recognition System

Radicioni

Esposito

2010

Advances in Music Information Retrieval

View full text Add to dashboard Cite

Tonal harmony analysis is a sophisticated task. It combines general knowledge with contextual cues, and it is concerned with faceted and evolving objects such as musical language, execution style and taste. We present BREVE, a system for performing a particular kind of harmony analysis, chord recognition: music is encoded as a sequence of sounding events and the system should assing the appropriate chord label to each event. The solution proposed to the problem relies on a conditional model, where domain knowledge is encoded in the form of Boolean features. BREVE exploits the recently proposed algorithm CarpeDiem to obtain significant computational gains in solving the optimization problem underlying the classification process. The implemented system has been validated on a corpus of chorales from J.S. Bach: we report and discuss the learnt weights, point out the committed errors, and elaborate on the correlation between errors and growth in the classification times in places where the music is less clearly asserted.

show abstract

“…Since harmony represents a significant part of what listeners respond to in music, its analysis is fundamental to a host of musical applications including expressive rendering, improvisational accompaniment systems, and may also constitute a one-dimensional reduction of music suitable for some search and retrieval applications. Past efforts in this area include (Pardo and Birmingham 2002), (Raphael and Stoddard 2003) (Temperley 2001), While the problem holds promise for a wide range of musical applications, the evaluation of such work remains difficult, primarily due to the scarcity of ground truth data as well as a suitable evaluation metric (not all errors are equally bad).…”

Section: Application To Harmonic Analysismentioning

confidence: 99%

Linear Dynamic Programming and the Training of Sequence Estimators

Raphael¹,

Nichols²

2009

Operations Research and Cyber-Infrastructure

Self Cite

View full text Add to dashboard Cite

We consider the problem of finding an optimal path through a trellis graph when the arc costs are linear functions of an unknown parameter vector. In this context we develop an algorithm, Linear Dynamic Programming (LDP), that simultaneously computes the optimal path for all values of the parameter. We show how the LDP algorithm can be used for supervised learning of the arc costs for a dynamic-programming-based sequence estimator by minimizing empirical risk. We present an application to musical harmonic analysis in which we optimize the performance of our estimator by seeking the parameter value generating the sequence best agreeing with hand-labeled data.

show abstract

Functional Harmonic Analysis Using Probabilistic Models

Cited by 75 publications

References 5 publications

Towards Cross-Version Harmonic Analysis of Music

Towards Cross-Version Harmonic Analysis of Music

BREVE: An HMPerceptron-Based Chord Recognition System

Linear Dynamic Programming and the Training of Sequence Estimators

Contact Info

Product

Resources

About