Rocco Zaccagnino

An Evolutionary Music Composer Algorithm for Bass Harmonization

In this paper we present an automatic Evolutionary Music Composer algorithm and a preliminary prototype software that implements it. The specific music composition problem that we consider is the so called unfigured (or figured) bass problem: a bass line is given (sometimes with information about the chords to use) and the automatic composer has to write other 3 voices to have a complete 4-voice piece of music. By automatic we mean that there must be no human intervention in the composing process. We use a genetic algorithm to tackle the figured bass problem and an ad-hoc algorithm to transform an unfigured bass to a figured bass. In this paper we focus on the genetic algorithm.

EvoBassComposer: a multi-objective genetic algorithm for 4-voice compositions

In this paper we consider the musical problem called unfigured bass harmonization: a bass line is given and the composer has to write other 3 voices to have a complete 4-voice piece of music with a 4-note chord for each bass note. Solving such a problem means finding appropriate chords to use for each bass note and also find a placement of the four notes within each chord so that melodic concerns are addressed, especially for the highest voice (soprano). We present a multi-objective genetic algorithm that automatically composes music when provided with a bass line input. The objectives considered are two: the harmonic objective (finding appropriate chords) and the melodic objective (find good melodic lines).

A neural network for bass functional harmonization

This paper presents the design, implementation and testing of a neural network for the functional harmonization of a bass line. The overall network consists of three base networks that are used in parallel under the control of an additional network that, at each step, chooses the best output from the three base networks. All the neural networks have been trained using J.S. Bachs chorales. In order to evaluate the networks, a metric measuring the distance of the output from the original J.S. Bachs harmonization is defined.

A hybrid computational intelligence approach for automatic music composition

The use of computers in the production of artifacts has drawn the attention of both artists and computer scientists. Among the different art disciplines, music is one of the arts that most benefited from the use of computers. There are many works which demonstrate the great synergy between these two fields. In this paper we will focus on a specific music composition problem: the figured bass problem, in which we have to automatically generate a 4 voice piece of music, starting from an input the bass line. To solve this problem we use a hybrid strategy, in which different metaheuristics cooperate to find high quality solutions. The cooperation is controlled by means of the combination of fuzzy control and knowledge obtained through Data Mining. As will be shown in the experimental results section, this hybrid strategy is capable of finding musical solutions with an acceptable quality and never discordant which, according to experts, are sound and adhere to scholastic rule.

A multi-objective differential evolution algorithm for 4-voice compositions

In this paper we explore the use of differential evolution for the musical problem called unfigured bass harmonization: a bass line is given and the composer has to write other 3 voices to have a complete 4-voice piece of music with a 4-note chord for each bass note. Solving such a problem means finding appropriate chords to use for each bass note and also find a placement of the four notes within each chord so that melodic concerns are addressed. Other papers have tackled this problem using evolutionary algorithms. In this paper we show how we can improve on these algorithms using differential evolution. In our algorithm we follow the standard differential evolution strategy but we also use customizations specific to the problem considered. We have implemented in Java our algorithm and we have run several tests to evaluate it. The results show that the differential approach allows us to improve on previous algorithms.

A genetic algorithm for dodecaphonic compositions

In this paper we propose an automatic music composition system for dodecaphonic music based on a genetic algorithm. Dodecaphonic music, introduced by A. Schoenberg, departs from the concept of tonality by considering all 12 notes equally important. Dodecaphonic compositions are constructed starting from a 12-note series, which is a sequence of all the 12 notes; the compositional process uses the starting 12-note series as a seed and builds the music creating new fragments of music obtained by transforming the seed series. The algorithm proposed in this paper automates the compositional process taking a seed series as input and automatically creating a dodecaphonic composition. We have implemented the algorithm and we have run several tests to study its behaviour.

Chorale Music Splicing System: An Algorithmic Music Composer Inspired by Molecular Splicing

Splicing systems are a formal model of a generative mechanism of words (strings of characters), inspired by a recombinant behavior of DNA. They are defined by a finite alphabet \(\mathcal{A}\), an initial set \(\mathcal{I}\) of words and a set \(\mathcal{R}\) of rules. Many of the studies about splicing systems focused on the properties of the generated languages and their theoretical computational power.

A Kind of Bio-inspired Learning of mUsic stylE

In the field of Computer Music, computational intelligence approaches are very relevant for music information retrieval applications. A challenging task in this area is the automatic recognition of musical styles. The style of a music performer is the result of the combination of several factors such as experience, personality, preferences, especially in music genres where the improvisation plays an important role.

Understanding the structure of musical compositions: Is visualization an effective approach?:

Experienced musicians have the ability to understand the structural elements of music compositions. Such an ability is built over time through the study of music theory, the understanding of rules that guide the composition of music, and countless hours of practice. The learning process is hard, especially for classical music, where the rigidity of the music structures and styles requires great effort to understand, assimilate, and then master the learned notions. In particular, we focused our attention on a specific type of music compositions, namely, music in chorale style (four-voice music). Composing such type of music is often perceived as a difficult task because of the rules the composer has to adhere to. In this article, we propose a visualization technique that can help people lacking a strong knowledge of music theory. The technique exploits graphic elements to draw the attention on the possible errors in the composition. We then developed an interactive system, named VisualMelody, that employs the proposed visualization technique to facilitate the understanding of the structure of music compositions. The aim is to allow people to make four-voice music composition in a quick and effective way, that is, avoiding errors, as dictated by classical music theory rules. We have involved 40 people in testing VisualMelody in order to analyze its effectiveness, its usability, and the overall user satisfaction. We partitioned the people involved in the evaluation study into two groups evenly splitting the musical expertise. Then, we had one group use VisualMelody without the visualization facilities and the other using the tool enhanced with our visualization. On average, people in the group that used our visualization were 60% faster and produced music with less errors.

Splicing music composition

Splicing systems were introduced by Tom Head (1987) as a formal model of a recombination process between DNA molecules. The existing literature on splicing systems mainly focuses on the computational power of these systems and on the properties of the generated languages; very few applications based on splicing systems have been introduced.In this paper we show a novel application of splicing systems: we use them to build an automatic music composer. The proposed system can be seen also as a new valid bio-inspired strategy for automatic music composition. It is tailored on 4-voice chorale-like music. We define a new music representation based on words, which extends an earlier splicing approach and uses additional music information to produce music in a quick and effective way.A performance study shows that our composer outperforms other meta-heuristics by producing better music according to a specific measure of quality evaluation. Moreover, the composition is carried out in a shorter time and using less memory with respect to a previous approach.