GAN-based Content-Conditioned Generation of Handwritten Musical Symbols

TL;DR

This paper introduces a GAN-based method to generate realistic handwritten musical symbols, aiming to improve optical music recognition by augmenting training data with synthetic, high-fidelity samples.

Contribution

It presents a novel music symbol-level GAN that produces realistic handwritten musical symbols and assembles them into full scores, advancing synthetic data generation for OMR.

Findings

Generated symbols show high visual realism

Synthetic scores can potentially enhance recognition models

Progress in realistic handwritten score synthesis

Abstract

The field of Optical Music Recognition (OMR) is currently hindered by the scarcity of real annotated data, particularly when dealing with handwritten historical musical scores. In similar fields, such as Handwritten Text Recognition, it was proven that synthetic examples produced with image generation techniques could help to train better-performing recognition architectures. This study explores the generation of realistic, handwritten-looking scores by implementing a music symbol-level Generative Adversarial Network (GAN) and assembling its output into a full score using the Smashcima engraving software. We have systematically evaluated the visual fidelity of these generated samples, concluding that the generated symbols exhibit a high degree of realism, marking significant progress in synthetic score generation.