The Beauty of Repetition in Machine Composition Scenarios

Abstract

Repetition, a basic form of artistic creation, appears in most musical works and delivers enthralling aesthetic experiences. However, repetition remains underexplored in terms of automatic music composition. As an initial effort in repetition modelling, this paper focuses on generating motif-level repetitions via domain knowledge-based and example-based learning techniques. A novel repetition transformer (R-Transformer) that combines a Transformer encoder and a repetition-aware learner is trained on a new repetition dataset with 584,329 samples from different categories of motif repetition. The Transformer encoder learns the representation among music notes from the repetition dataset; the novel repetition-aware learner exploits repetitions' unique characteristics based on music theory. Experiments show that, with any given motif, R-Transformer can generate a large number of variable and beautiful repetitions. With ingenious fusion of these high-quality pieces, the musicality and appeal of machine-composed music have been greatly improved. CCS CONCEPTS• Applied computing → Sound and music computing.

Music and Mathematics: Processing Repetition Techniques with the Least Common Multiple (LCM) in Rajékan Music Composition

“…A type of music with repetition considered to work effectively "hypnotizing" to bring game players into the desired mindset towards gaming, Rajékan is one part of manipulation the listener through the LCM technique. (Hu et al, 2022).…”

Responding to the Call: Exploring Automatic Music Composition Using a Knowledge-Enhanced Model

“…We use dynamic time warping (DTW) distance and maximum common subsequence (MCS). These metrics have been widely used in music composition for similarity comparisons (Collins 2012;Dai et al 2023;Chikkamath and Nirmala 2021;Hu et al 2022).…”

“…We carry out a subjective evaluation on social media. The listener evaluates the music in terms of quality, groove, coherence, and creativity (Zhang et al 2022;Hu et al 2022): Overall Quality: Please rate the overall quality of the response on a scale of 1 (lowest) to 5 (highest). Groove: Please rate the fluency of the response and the appropriateness of pauses on a scale of 1 (lowest) to 5 (highest).…”