Symbolic Music Similarity through a Graph-Based Representation

Abstract

In this work, a novel representation system for symbolic music is described. The proposed representation system is graph-based and could theoretically represent music both from a horizontal (contrapuntal) and from a vertical (harmonic) point of view, by keeping into account contextual and harmonic information. It could also include relationships between internal variations of motifs and themes. This is achieved by gradually simplifying the melodies and generating layers of reductions that include only the most important notes from a structural and harmonic viewpoint. This representation system has been tested in a music information retrieval task, namely melodic similarity, and compared to another system that performs the same task but does not consider any contextual or harmonic information, showing how the structural information is needed in order to find certain relations between musical pieces. Moreover, a new dataset consisting of more than 5000 leadsheets is presented, with additional meta-musical information taken from different web databases, including author, year of first performance, lyrics, genre and stylistic tags. CCS CONCEPTS• Information systems → Music retrieval; • Applied computing → Sound and music computing;

A rhythm-aware serious game for social interaction

“…It would be useful to assess if using user-selected melodic seeds can make the experience more enjoyable to users. The chosen melodies should also be able to adapt to the changes in the input, possibly defining more hierarchical levels of different note densities [44] and embedding expressive performance generation to leverage such hierarchical descriptions [13,18].…”

A Framework for Content-Based Search in Large Music Collections

“…This draws heavily from [25][26][27], which introduced into the music information retrieval literature some ideas and tools from the fields of databases systems and computer linguistics (e.g., hierarchical decomposition of musical content and context-free grammars). Recently, the authors of [28] used a similar graph-based representation to study music similarity.…”

Computational Creativity and Music Generation Systems: An Introduction to the State of the Art

“…Each of these steps follows a different algorithm, but some information is passed on through each step. In particular, all of the steps use information about the importance of each of the generated notes, dividing them with a Schenkerian approach (Simonetta et al, 2018 ). The authors argue that this idea can be further extended to other tasks (such as form generation and harmony), including both deep learning and classical AI algorithms, trying to find the optimal combination for each task (Carnovalini, 2019 ).…”