Synthetic Music Random Generation Based on Nyquist

Abstract

The idea of composing music based on programs has a long history, from some pioneers who first tried to fulfill this idea, to now, the variety of software and workshops that allow people to generate music in different approaches. On this basis, this study will briefly discuss the main idea and some methods used to compose the music in this paper. We aimed to randomly generate a piece of music using only Nyquist. Thus, each time when listen to this, there is a slight difference. With this in mind, this paper has used two ways: the first one is to use several functions in Nyquist, which would randomly choose numbers in a given list or range. The second way is a method based on the functions mentioned earlier, called the random walk, which lets the program stochastically choose one item in a given list of a range of numbers every time to add or subtract to the origin number given by programmers. By using these two techniques, one can generate random sounds and random rhythmic patterns, and when we put these sounds and patterns into scores to generate various melodies, we can stochastically increase or decrease the overall pitch of that score. Finally, the scores are combined to create the whole piece of music. Although there are limitations, this study has provided some methods that can be useful for generating stochastic music.