A Deeper Look at Sheet Music Composer Classification Using Self-Supervised Pretraining

• Main Authors: Daniel Yang, Kevin Ji, TJ Tsai
• Format: Article
• Language: English
• Published: MDPI AG 2021-02-01
• Series: Applied Sciences
• Subjects: sheet music; style recognition; composer identification; language model; pretraining; self-supervised
• Online Access: https://www.mdpi.com/2076-3417/11/4/1387

This article studies a composer style classification task based on raw sheet music images. While previous works on composer recognition have relied exclusively on supervised learning, we explore the use of self-supervised pretraining methods that have been recently developed for natural language processing. We first convert sheet music images to sequences of musical words, train a language model on a large set of unlabeled musical “sentences”, initialize a classifier with the pretrained language model weights, and then finetune the classifier on a small set of labeled data. We conduct extensive experiments on International Music Score Library Project (IMSLP) piano data using a range of modern language model architectures. We show that pretraining substantially improves classification performance and that Transformer-based architectures perform best. We also introduce two data augmentation strategies and present evidence that the model learns generalizable and semantically meaningful information.