On large-scale genre classification in symbolically encoded music by automatic identification of repeating patterns

Abstract

The importance of repetitions in music is well-known. In this paper, we study music repetitions in the context of effective and efficient automatic genre classification in large-scale music-databases. We aim at enhancing the access and organization of pieces of music in Digital Libraries by allowing automatic categorization of entire collections by considering only their musical content. We handover to the public a set of genre-specific patterns to support research in musicology. The patterns can be used, for instance, to explore and analyze the relations between musical genres.There are many existing algorithms that could be used to identify and extract repeating patterns in symbolically encoded music. In our case, the extracted patterns are used as representations of the pieces of music on the underlying corpus and, consecutively, to train and evaluate a classifier to automatically identify genres. In this paper, we apply two very fast algorithms enabling us to experiment on large and diverse corpora. Thus, we are able to find patterns with strong discrimination power that can be used in various applications. We carried out experiments on a corpus containing over 40,000 MIDI files annotated with at least one genre. The experiments suggest that our approach is scalable and capable of dealing with real-world-size music collections.

Transposition and time-warp invariant algorithm for detecting repeated patterns in polyphonic music

“…The algorithm finds for each input note pair the longest repeating pattern whose last note pair is that pair. We also show how the algorithm can be extended (4,4), and (6,5), and the second pattern consists of notes (5, 1), (7,3), (8,2), and (10,3). The intervals between consecutive notes are the same, but the onset time differences are different.…”

“…. ., (x m , y m ) where The first pattern consists of notes (2, 2), (4, 4) and (5,3), and the second pattern consists of notes (2,4), (3,6) and (6,5). The interval of each note pair is 2.…”

“…This makes it easier for listeners to detect musical ideas and remember music [9]. Found repetitions can also be used for genre classification purposes [2].…”

MidiBERT-Piano: Large-scale Pre-training for Symbolic Music Understanding

“…Second, we aim at establishing a benchmark for symbolic music understanding and include not only sequencelevel but also note-level tasks. Moreover, the sizes of the labeled data for our downstream tasks are all almost under 1K, while that employed in MusicBERT (i.e., the TOP-MAGD dataset [58]) contains over 20K annotated pieces, a data size rarely seen for symbolic music tasks in general. Finally, while their token representation is designed for multi-track MIDI, ours is for single-track piano scores.…”

MusicBERT: Symbolic Music Understanding with Large-Scale Pre-Training

“…Genre classification and style classification (Ferraro and Lemström, 2018) are multi-label classification tasks. Following Ferraro and Lemström (2018), we use the TOP-MAGD dataset for genre classification and the MASD dataset for style classification. TOP-MAGD contains 22,535 annotated files of 13 genres, and MASD contains 17,785 files of 25 styles.…”