Monophonic music composition using genetic algorithm and Bresenham's line algorithm

Music composition is one of the oldest artistic pursuits. The role of machine in automatic generation of creative artworks, like music, is still an explorable area. In this paper, a new approach for music composition is proposed that differs from previous methods of generating music with predefined musical parameters. The proposed method gives flexibility to change the musical parameters, namely time signature, range and scale. Moreover, the user can also input any desired motif (a short piece of melody) and generate a melody that includes the motif. The type of music composed in this work is monophonic, which includes melody and rhythm. To create a proper sequence of notes, we used genetic algorithm with suitably formulated crossover and mutation operators. The rhythm is generated using Bresenham's line drawing algorithm, which has been modified to adapt different time signature with changing beats. PySynth, a simple music synthesizer, is chosen to convert music into wave file format. In the end, a comparative analysis is conducted to show the efficacy of the proposed model. Results show that the proposed algorithm performs better than conventional genetic algorithm.