• Corpus ID: 2600270

Towards a Computational Model of Melody Identification in Polyphonic Music

  title={Towards a Computational Model of Melody Identification in Polyphonic Music},
  author={S{\o}ren Tjagvad Madsen and Gerhard Widmer},
This paper presents first steps towards a simple, robust computational model of automatic melody identification. Based on results from music psychology that indicate a relationship between melodic complexity and a listener's attention, we postulate a relationship between musical complexity and the probability of a musical line to be perceived as the melody. We introduce a simple measure of melodic complexity, present an algorithm for predicting the most likely melody note at any point in a… 

Figures and Tables from this paper

A Complexity-based Approach to Melody Track Identification in MIDI Files ?
The importance of music com- plexity as a factor for melody recognition in multi-voiced popular music is examined, and measures of melodic complexity calculated from pitch and timing information are proposed.
What Defines a Melody? A MIDI Musical Line Classification on Features Relative from the Musical Line to the Musical Piece
In order to test which musical aspects relate to what defines the melody in a musical piece, a method for feature selection, and an implementation for phrase segmentation of musical pieces have been
Towards a Human-Friendly Melody Characterization by Automatically Induced Rules
The goal of this work is to analyse how statistical rules can be used to characterize a melody in such a way that one can understand the solution of an automatic system for selecting the track containing the melody in standard MIDI file format.
Feature Extraction Using Pitch Class Profile Information Entropy
This work extends ideas that estimate information entropy of sections of musical pieces, to utilize the Pitch Class Profile information entropy for global feature extraction, and proposes and tests two approaches.
Effects of Repetition on Attention in Two-Part Counterpoint
Repetition and novelty are essential components of tonal music. Previous research suggests that the degree of repetitiveness of a line can determine its relative melodicity within a musical texture.
Numerical Tools for Describing Musical Compositions
Our goal for this paper is to apply the methodologies of statistical mechanics to develop exible tools for use in computational musicology. Classical statistical mechanics is the study of extremely
YQX Plays Chopin
The article is about AI research in the context of a complex artistic behavior: expressive music performance. A computer program is presented that learns to play piano with 'expression' and that even
Related Work and a Taxonomy of Musical Intelligence Tasks
  • Elad Liebman
  • Computer Science
    Sequential Decision-Making in Musical Intelligence
  • 2019
This section will not only introduce a broad overview of the related literature, but also propose a unified framework for parsing this varied and complex body of work, delineating a taxonomy for music AI tasks, and mapping out the overall state of the art.
Melodic Track Identification in MIDI Files Considering the Imbalanced Context
Experimental results over three different music genres prove that learning from balanced training sets clearly provides better results than the standard classification process.
Artificial Musical Intelligence: A Survey
A definition of musical intelligence is provided, a taxonomy of its constituent components is introduced, and the wide range of AI methods that can be, and have been, brought to bear in its pursuit are surveyed, with a particular emphasis on machine learning methods.


Melodic analysis with segment classes
A representation for melodic segment classes is presented and it is applied to sequential pattern discovery and to the statistical modeling of musical style.
A Musical Pattern Discovery System Founded on a Modeling of Listening Strategies
This chapter discusses motivic analysis as a set of partially detailed operations that carry out a top–down hierarchical segmentation of the musical work, which limits a detailed understanding of musical structure.
Music and Memory: An Introduction
This far-ranging book shows how human memory influences the organization of music. The book is divided into two parts. The first part presents basic ideas about memory and perception from cognitive
Manipulation of music for melody matching
This paper analyzes the propeties of music, music perception, and musicdatabase users, and uses the analysis to propose alternative techniques for extracting monophonic melodies from polyphonic music; it is believed that such melodies can subsequently be used formatching of queries to data.
Structural and affective aspects of music from statistical audio signal analysis
The analysis shows strong evidence that signal properties and human reactions are related, suggesting applications of these techniques to music understanding and music information-retrieval systems.
Structural and affective aspects of music from statistical audio signal analysis: Special Topic Section on Computational Analysis of Style
Understanding and modeling human experience and emotional response when listening to music are important for better understanding of the stylistic choices in musical composition. In this work, we
Melodic track identification in MIDI files
This work developed a random forest classifier system that assigns a probability of being a melodic line to each track of the target melody, and results were obtained testing different MIDI databases.
Efficient Melody Retrieval with Motif Contour Classes
Empirical and analytical results on the retrieval quality, the complexity, and quality/efficiency trade-off in the use of motif contour classes for efficient retrieval of melodies from music collections are presented.
The Performance of Music
Separating voices in MIDI
An algorithm for converting midi events into logical voices based on the pitch proximity principle is presented and new heuristics are introduced and evaluated in order to handle unsolved situations.