Conceptualising a theoretical device to synthesise melodies.

https://www.scirp.org/html/2224.html

Revolves around talk of AI for musical synthesis.

Foreword. #

Algorithmic Music Construction provides a logical framework for the theoretical generation of human melodies & overall structure of music.

Their method included four stages:

1. Specification of the theoretical basis
2. Translation of the theory into a set of principles
3. Implementation of the principles as a generative computer algorithm
4. Test of the output.

Various methodology of creating algorithmic composition includes Markov chains, Knowledge-based systems, Grammars, Evolutionary methods divided into genetic algorithms, & Interactive genetic algorithms.

Because the method presented in the paper generates music, it can be seen as an instance of algorithmic composition. However, since it was not meant as a test on musical theory rather than classical AI methods, to guide the implementation of the music generating device.

Base model #

The starting point of the project involves the conception of music as a psychological phenomenon^[1]; in the cases of tonal music this would involve the combination of two distinct aspects found within its methodology;

1. Discovering the context of which the music was conceived; i.e: its meter^[2], key, harmony, rhythm etc, and to represent the input which consists in sounds of varying pitch. This context is then used to infer its intent & emotion, as these now form the musical themes & motifs commonly associated with the representations of moods within humans.^[2:1]
2. Within that context, tone sequences are generated using construction rules specifying the hierarchical organisation of tones into parts, & of parts into larger parts, etc., relating to concepts like motives, phrases, repetition & variation, skeleton, structure & ornamental tones, etc. In line with these assumptions, two components are discerned from

Go back