C++, neural networks and fuzzy logic (2nd ed.)
C++, neural networks and fuzzy logic (2nd ed.)
Genetic algorithms + data structures = evolution programs (3rd ed.)
Genetic algorithms + data structures = evolution programs (3rd ed.)
Genetic Algorithms in Search, Optimization and Machine Learning
Genetic Algorithms in Search, Optimization and Machine Learning
Fuzzy and Neural Approaches in Engineering
Fuzzy and Neural Approaches in Engineering
IEEE Transactions on Consumer Electronics
Hi-index | 0.00 |
A Genetic Algorithm is used in this paper to compose music exhibiting the characteristics of traditional Greek music, and especially Greek dances. Since the scales and the intonation phenomena of the harmonic unisons involved with the sample space are in many aspects incompatible with Western music, divergence and de-escalation are provoked. In order to enhance the standard MIDI protocol with Delta transitory phenomena that have been observed in Greek music forms, fuzzy transformations are proposed. Rules produced from them are fed into a GA to compose music. The GA has the ability to explore a vast region of the composition space through its mutation parameters. At the same time the algorithm is carefully directed, through its fitness function, to observe the composition rules of the prototyped melodies. These features make the algorithm a good tool for simulating the procedure of improvisation. The fitness function of the GA is derived from the modal characteristics (our rules) evolving from the knowledge base of the context of the prototyped melodic lines, so the resulting music is bounded by the main characteristics of the prototype in terms of scales, rhythm and thematic inclinations. The proposed algorithm is transformative and generative as well.