Large-Scale MIDI-based Composer Classification

TL;DR

This paper introduces a large-scale MIDI-based system for composer classification using deep neural networks, achieving notable accuracy improvements over audio-based methods on datasets with up to 100 composers.

Contribution

It is the first to explore composer classification with up to 100 composers using MIDI data and deep neural networks, demonstrating the effectiveness of MIDI representations.

Findings

Achieved 0.648 accuracy for 10 composers on 30-second clips

Achieved 0.385 accuracy for 100 composers on 30-second clips

Outperformed audio-based baseline systems in composer classification

Abstract

Music classification is a task to classify a music piece into labels such as genres or composers. We propose large-scale MIDI based composer classification systems using GiantMIDI-Piano, a transcription-based dataset. We propose to use piano rolls, onset rolls, and velocity rolls as input representations and use deep neural networks as classifiers. To our knowledge, we are the first to investigate the composer classification problem with up to 100 composers. By using convolutional recurrent neural networks as models, our MIDI based composer classification system achieves a 10-composer and a 100-composer classification accuracies of 0.648 and 0.385 (evaluated on 30-second clips) and 0.739 and 0.489 (evaluated on music pieces), respectively. Our MIDI based composer system outperforms several audio-based baseline classification systems, indicating the effectiveness of using compact MIDI representations for composer classification.