Emilios Cambouropoulos

• PhD
• Professor at Aristotle University of Thessaloniki

Recent advances in music source separation through deep learning algorithms have enabled the isolation of vocals from recordings where separate tracks are not available. We have used Demucs (v4) Music Source Separation (Défossez, 2021; Rouard et al., 2023) to obtain isolated vocal parts from song recordings by Vasilis Tsitsanis, a prominent songwri...

Stelios Kazantzidis is one of the most prominent singers of popular Greek music with a career spanning from the 1950s till the end of the 20th century. His collaboration with the iconic composer Vassilis Tsitsanis, shaped the post-war Greek music scene and he later collaborated with most of the prominent Greek composers of popular music. One of his...

Η μουσική παίζει σημαντικό ρόλο στη ζωή του ανθρώπου, από το πρώτα κιόλας στάδια, όντας άρρηκτα συνδεδεμένη με τα συναισθήματα (Swaminathan & Schellenberg, 2015). Η σχέση των συναισθημάτων με τη μουσική μπορεί να διακριθεί σε δύο διαφορετικά επίπεδα: στα συναισθήματα που οι ακροατές αντιλαμβάνονται ότι η μουσική προσπαθεί να εκφράσει (perceived emo...

In this study we look into how repeated listening to a classical music excerpt may affect the perception of its structure and to what extent types of repetition may shape the perception of musical entities. Thirty-eight participants with musical training were asked to indicate in real-time the points where they identified the introduction of 'music...

This paper presents a methodology for generating cross-harmonizations of jazz standards, i.e., for harmonizing the melody of a jazz standard (Song A) with the harmonic context of another (Song B). Specifically, the melody of Song A, along with the chords that start and end its sections (chord constraints), are used as a basis for generating new har...

This paper presents current developments taking place in the context of the MusiCoLab project. MusiCoLab aims at delivering a comprehensive and efficient web platform for music learning and teaching, by building on prior research experience of project partners, as well as by investigating the integration of state-of-the-art tools in intelligent mus...

This chapter discusses musical similarity focusing on issues of representation and processing of patterns in symbolic music data. Various facets of musical similarity are explored that pertain to practical problems encountered when developing formal models for pattern identification and induction in musical corpora; the representation of musical da...

Visual associations with auditory stimuli have been the subject of numerous studies. Colour, shape, size, and several other parameters have been linked to musical elements like timbre and pitch. In this article, we aim to examine the relationship between harmonisations with varying degrees of dissonance and visual roughness. Based on past research...

Encoding note simultaneities (chords) has been approached in different ways, such as Roman numerals for tonal harmony, or pitch class sets encountered in atonal and non-tonal music. A novel chord representation, the General Chord Type (GCT) representation, was developed to be adaptable to a broad variety of harmonic idioms from tonal to atonal. Giv...

In this chapter, recent research in the domain of melodic harmonization and computational creativity is presented with a view to highlighting strengths and weaknesses of the classical cognitively inspired symbolic AI approach (often in juxtaposition to contemporary deep learning methodologies). A modular melodic harmonization system that learns cho...

The difficulty of modelling musical structure in a general and cognitively plausible manner is due primarily to music’s inter-dependent multi-parametric and multi-level nature that allows multiple structural interpretations to emerge. Traditional AI symbolic processing methods, however, can be used effectively for modelling particular analytic and...

CHAMELEON is a computational melodic harmonization assistant. It can harmonize a given melody according to a number of independent harmonic idioms or blends between idioms based on principles of conceptual blending theory. Thus, the system is capable of offering a wealth of possible solutions and viewpoints for melodic harmonization. This study inv...

Visual associations with auditory stimuli have been the subject of numerous studies. Colour, shape, size, and several other parameters have been linked to musical elements like timbre and pitch. In this paper we aim to examine the relationship between harmonisations with varying degrees of dissonance and visual roughness. Based on past research in...

This paper presents an experiment designed to investigate the influence of a creativity support tool on music creation. Twenty five participants were asked to harmonise two very similar melodies, the first on their own and the second while given the opportunity to interact with the CHAMELEON harmonisation assistant. CHAMELEON can offer a variety of...

This study explores the relationship between music and movement, focusing on hand movements in relation to electronically produced sound events (granular synthesis). This relation is studied empirically by presenting pairs of hand and sound gestures(in the form of videos)to participants, while trying to find cases where correlations exist between t...

Conceptual Blending (CB) theory describes the cognitive mechanisms underlying the way humans process the emergence of new conceptual spaces by blending two input spaces. CB theory has been primarily used as a method for interpreting creative artefacts, while recently it has been utilised in the context of computational creativity for algorithmic in...

This article presents the CHAMELEON melodic harmonisation assistant that utilises the principles of conceptual blending theory as a means for the invention of hybrid or novel harmonic idioms and an empirical evaluation of a number of computer-generated melodic harmonisation blends. Melodies originating from various idioms were harmonised either acc...

This chapter presents the CHAMELEON melodic harmonisation assistant that learns different aspects of harmony from expert-annotated data, blends learnt harmonies from different idioms using the COINVENT framework and harmonises user-given melodies. The learnt harmonic elements include chord types, chord transitions, cadences and bass voice leading,...

The concept of root is of great significance in chord encoding in tonal music. Is this notion useful in non-tonal idioms or should it be extended, changed or abandoned in different musical contexts? A series of harmonic excerpts from diverse idioms are examined through the application of different root-finding and chord encoding models, such as Par...

In computational creativity, new concepts can be invented through conceptual blending of two independent conceptual spaces. In music, conceptual blending has been primarily used for analysing relations between musical and extra-musical elements in composed music rather than generating new music. This paper presents a probabilistic melodic harmonisa...

THE COGNITIVE THEORY OF CONCEPTUAL BLENDING may be employed to understand the way music becomes meaningful and, at the same time, it may form a basis for musical creativity per se. This work constitutes a case study whereby conceptual blending is used as a creative tool for inventing musical cadences. Specifically, the perfect and the renaissance P...

Conceptual blending when used as a creative tool combines the features of two input spaces, generating new blended spaces that share the common structure of the inputs, as well as different combinations of their non-common parts. In the case of music, conceptual blending has been employed creatively, among others, in generating new cadences (pairs...

How can harmony in diverse idioms be represented in a machine learning system and how can learned harmonic descriptions of two musical idioms be blended to create new ones? This paper presents a creative melodic harmonisation assistant that employs statistical learning to learn harmonies from human annotated data in practically any style, blends th...

Human music listeners are capable of identifying multiple `voices' in musical content. This capability of grouping notes of polyphonic musical content into entities is of great importance for numerous processes of the Music Information Research domain, most notably for the better understanding of the underlying musical content's score. Accordingly,...

Melodic harmonisation is a sophisticated creative process that involves deep musical understanding and a specialised knowledge of music relating to melodic structure, harmony, rhythm, texture, and form. In this article a new melodic harmonisation assistant is presented that is adaptive (learns from data), general (can cope with any tonal or non-ton...

Conceptual blending is a powerful tool for computational creativity where, for example, the properties of two harmonic spaces may be combined in a consistent manner to produce a novel harmonic space. However, deciding about the importance of property features in the input spaces and evaluating the results of conceptual blending is a nontrivial task...

Selecting an appropriate representation for chords is important for encoding pertinent harmonic aspects of the musical surface, and, at the same time, is crucial for building effective computational models for music analysis. This chapter, initially, addresses musicological, perceptual and computational aspects of the harmonic musical surface. Then...

This paper presents two novel strategies for processing chroma vectors corresponding to polyphonic audio, and producing a symbolic representation known as GCT (General Chord Type). This corresponds to a fundamental step in the conversion of general polyphonic audio files to this symbolic representation, which is required for enlarging the current c...

In this paper we critically investigate the application of Fauconnier & Turner's Conceptual Blending Theory (CBT) in music, to expose a series of questions and aporias highlighted by current and recent theoretical work in the field. Investigating divisions between different levels of musical conceptualization and blending, we question the common di...

Melodic harmonisation deals with the assignment of harmony (chords) over a given melody. Probabilistic approaches to melodic harmonisation utilise statistical information derived from a training dataset to harmonise a melody. This paper proposes a probabilistic approach for the automatic generation of voice leading for the bass note on a set of giv...

We present a computational framework for chord invention based on a cognitive-theoretic perspective on conceptual blending. The framework builds on algebraic specifications, and solves two musicological problems. It automatically finds transitions between chord progressions of different keys or idioms, and it substitutes chords in a chord progressi...

This research focuses on concept invention processes and suggests that structural blending is a powerful mechanism that gives rise to novel musical concepts. Structural blending is omnipresent across several formal musical levels, from individual pieces harmoniously combining music characteristics of different pieces/styles, to entire musical style...

Conceptual blending is a cognitive theory whereby elements from diverse, but structurally-related, mental spaces are 'blended' giving rise to new conceptual spaces. This study focuses on structural blending utilising an algorith-mic formalisation for conceptual blending applied to harmonic concepts. More specifically, it investigates the ability of...

Probabilistic methodologies provide successful tools for automated music composition, such as melodic harmoni-sation, since they capture statistical rules of the music idioms they are trained with. Proposed methodologies focus either on specific aspects of harmony (e.g., generating abstract chord symbols) or incorporate the determination of many ha...

The General Chord Type (GCT) representation is appropriate for encoding tone simultaneities in any harmonic context (such as tonal, modal, jazz, octatonic, atonal). The GCT allows the rearrangement of the notes of a harmonic sonority such that abstract idiom-specific types of chords may be derived. This encoding is inspired by the standard roman nu...

Conceptual blending is a cognitive theory whereby elements from diverse, but structurally-related, mental spaces are " blended " giving rise to new conceptual spaces that often possess new powerful interpretative properties, allowing better understanding of known concepts or the emergence of novel concepts altogether. This paper provides an overvie...

In this paper we focus on issues of harmonic representa-tion and computational analysis. A new idiom-independent representation is proposed of chord types that is appropriate for encoding tone simultaneities in any harmonic context (such as tonal, modal, jazz, octatonic, atonal). The General Chord Type (GCT) representation, allows the re-arrangemen...

During the last decades, several methodologies have been proposed for the harmonization of a given melody with al-gorithmic means. Among the most successful are method-ologies that incorporate probabilistic mechanisms and sta-tistical learning, since they have the ability to generate har-monies that statistically adhere to the harmonic character-is...

This paper examines a previously unstudied musical corpus de-rived from the polyphonic singing tradition of Epirus employing statistical methods. This analysis will mainly focus on unique harmonic aspects of these songs, which feature, for instance, unresolved dissonances (major second and minor seventh inter-vals) at structurally stable positions...

Composers commonly employ ornamentation and elaboration techniques to generate varied versions of an initial core melodic idea. Dynamic programming techniques, based on edit operations, are used to find similarities between melodic strings. However, replacements, insertions and deletions may give non-musically pertinent similarities, especially if...

In this paper, we address the problem of melodic string matching that enables identification of varied (ornamented) instances of a given melodic pattern. To this aim, a new set of edit distance operations adequate for pitch interval strings is introduced. Insertion, deletion and replacement operations are abolished as irrelevant. Consolidation and...

This paper addresses problems and misconceptions pertaining to the notion of the musical surface, a notion that is commonly thought to be relatively straight-forward and is often taken as a given in computational and cognitive research. It is suggested that the musical surface is comprised of (complex) musical events perceived as wholes within cohe...

In the first part of the paper a theoretical discussion is presented regarding the fundamental concept of similarity and its relation to cue abstraction and categorisation. It is maintained that similarity is by definition context-dependent and strongly interrelated to cue abstraction and categorisation. Emphasis is given to determining the “musica...

The Voice Integration/Segregation Algorithm (VISA) proposed by Karydis et al. [7] splits musical scores (symbolic musical data) into different voices, based on a perceptual view of musical voice that corresponds to the notion of auditory stream. A single ‘voice’ may consist of more than one synchronous notes that are perceived as belonging to the s...

Listeners are thought to be capable of perceiving multiple voices in music. This paper presents different views of what 'voice' means and how the problem of voice separation can be systematically described, with a view to understanding the problem better and developing a systematic description of the cognitive task of segregating voices in music. W...

A listener is thought to be able to organise musical notes into groups within musical streams/voices. A stream segment is a relatively short coherent sequence of tones that is separated horizontally from co-sounding streams and, vertically from neighbouring musical sequences. This paper presents a novel algorithm that discovers musical stream segme...

This paper proposes an efficient pattern extraction algorithm that can be applied on melodic sequences that are repre- sented as strings of abstract intervallic symbols; the melodic representation introduces special ''binary don't care'' symbols for intervals that may belong to two partially overlapping intervallic categories. As a special case the...

Listeners are capable to perceive multiple voices in music. Adopting a perceptual view of musical 'voice' that corresponds to the notion of auditory stream, a computational model is developed that splits musical scores (symbolic musical data) into different voices. A single 'voice' may consist of more than one synchronous notes that are perceived a...

Listeners are thought to be capable of perceiving multiple voices in music. Adopting a perceptual view of musical 'voice' that corresponds to the notion of auditory stream, a computational model is developed that splits a musical score (symbolic musical data) into different voices. A single 'voice' may consist of more than one synchronous notes tha...

We report three experiments examining the perception of tempo in expressively performed classical piano music. Each experiment investigates beat and tempo perception in a different way: rating the correspondence of a click track to a musical excerpt with which it was simultaneously presented; graphically marking the positions of the beats using an...

Despite the consideration that musical parallelism is an important factor for musical segmentation, there have been relatively few systematic attempts to describe exactly how it affects grouping processes. The main problem is that musical parallelism itself is difficult to formalize. In this study, a computational model that extracts melodic patter...

This paper proposes an efficient pattern extraction al- gorithm that can be applied on melodic sequences that are represented as strings of abstract intervallic symbols ; the melodic representation introduces special "don't care " symbols for intervals that may belong to two partially overlapping intervallic categories. As a special case the well e...

In this study we investigate the relationship between beat and musical performance. It is hypothesised that listeners prefer beat sequences that are smoother than beat tracks that are fully aligned with the actual onsets of performed notes. In order to examine this hypothesis, an experiment was designed whereby six different smoothed beat tracks ge...

In this article, cognitive and musicological aspects of pitch and pitch interval representations are explored via computational modeling. The specific task under investigation is pitch spelling, that is, how traditional score notation can be derived from a simple unstructured 12-tone repre- sentation (e.g., pitch-class set or MIDI pitch representat...

Despite the fact that musical parallelism is considered as an important factor for musical segmentation, there have been very few attempts to describe systematically how exactly it affects grouping processes. The main problem is that musical parallelism itself is very difficult to formalise. In this paper a computational model will be presented tha...

In order to analyse timing in musical performance, it is necessary to develop reliable and ecient methods of deriving musical timing information (e.g. tempo, beat and rhythm) from the physical timing of audio signals or MIDI data. We report the results of an experiment in which subjects were asked to mark the positions of beats in musical excerpts,...

Here we introduce two new notions of approximate matching with application in computer assisted music analysis. We present algorithms for each notion of approximation: for approximate string matching and for computing approximate squares.

In the first part of this article, the notions of identity, similarity, categorization, and feature salience are explored; musical examples are provided at various stages of the discussion. Then, formal working definitions are proposed that inextricably bind these concepts together. These definitions readily lend themselves to the development of a...

In this paper a number of issues relating to the application of string processing techniques on musical sequences are discussed. A brief survey of some musical string processing algorithms is given and some issues of melodic representation, abstraction, segmentation and categorisation are presented. This paper is not intended towards providing solu...

In this paper two main topics are addressed. Firstly, the Local Boundary Detection Model (LBDM) is described; this computational model enables the detection of local boundaries in a melodic surface and can be used for musical segmentation. The proposed model is tested against the punctuation rule system developed by Friberg et al. (1998) at KTH, St...

In this paper a computational model will be presented that attempts to organise melodic segments into ‘significant’ musical categories (e.g., motives). Given a segmentation of a melodic surface, the proposed algorithm constructs an appropriate representation for each segment in terms of a number of attributes (these reflect melodic and rhythmic asp...

. When a person taps a foot in time with a piece of music, they are performing beat tracking. Beat tracking is fundamental to the understanding of musical structure, and therefore an essential ability for any system which purports to exhibit musical intelligence or understanding. We present an off-line multiple agent beat tracking system which esti...

In this paper a formal model will be presented that attempts to organise melodic segments into `significant' musical categories (e.g. motives). Given a segmentation of a melodic surface, the proposed model constructs an appropriate representation for each segment in terms of a number of attributes (these reflect melodic and rhythmic aspects of the...

In this paper a number of issues relating to the application of string processing techniques on musical sequences are discussed. Special attention is given to musical pattern extraction. Firstly, a number of general problems are presented in terms of musical representation and pattern processing methodologies. Then a number of interesting melodic p...

this paper `properties' are taken to mean `binary features'

In this paper a system that is designed to extract the musical score from a MIDI performance is described. The proposed system comprises of a number of modules that perform the following tasks: identification of elementary musical objects, calculation of accent (salience) of musical events, beat induction, beat tracking, onset quantisation, streami...

Pitch and pitch‐intervals are most often represented — in the western tradition — either by the traditional pitch naming system and the relating pitch‐interval names, or as pitch‐classes and pitch‐class intervals. In this paper we discuss the properties, relationships and limits of these two representations and propose a General Pitch Interval Repr...

In this paper a general theory will be introduced that allows the description of a melodic surface — i.e. a score-like representation of a melody — in terms of local grouping, accentuation and metrical structures. Firstly, a formal model will be proposed that detects points of maximum local change that allow a listener to identify local perceptual...

In this paper a general theory will be introduced that facilitates the detection of local boundaries in melodic surfaces. A model will be proposed that discovers points of maximum local change that allow a listener to identify potential perceptual boundaries. It is suggested that the Local Boundary Detection Model (LBDM) presents a more effective m...

In this paper a formal model will be presented that attempts to organise melodic segments into 'significant' musical categories (e.g. motives). Given a segmentation of a melodic surface, the proposed model constructs an appropriate representation for each segment in terms of a number of attributes (these reflect melodic and rhythmic aspects of the...

In this paper a computational model is described that transcribes polyphonic MIDI pitch files into the Western traditional music notation. Input to the proposed algorithm input is merely a sequence of MIDI pitch numbers in the order they appear in a MIDI file. No a priori knowledge is required such as key signature, tonal centers, time signature, v...

In this paper a computational model is presented that extracts patterns from a given melodic surface and, then, following the assumption that the beginning and ending points of 'significant' repeating musical patterns influence the segmentation of a musical surface, the discovered patterns are used as a means to determine probable segmentation poin...

The notions of 'voice', as well as, homophony and polyphony, are thought to be well understood by musicians. Listeners are thought to be capable of perceiving multiple 'voices' in music. However, there exists no systematic theory that describes how 'voices' can be identified, especially, when polyphonic and homophonic elements are mixed together. T...

Background in music theory/analysis. Musical analysis focuses primarily on aspects of compositional design, mathematical/formal relations between musical materials or on musical theoretic forms and functions that have been established as musicologically pertinent through the centuries (e.g., traditional harmonic analysis, Schenkerian analysis). Lis...