Bach or Mock? A Grading Function for Chorales in the Style of J.S. Bach

TL;DR

This paper introduces an interpretable grading function for evaluating Bach-style chorales, which effectively distinguishes between real and generated music and outperforms human experts in this task.

Contribution

The paper presents a novel, musically-motivated grading function for evaluating four-part chorales in Bach's style, applicable to deep generative models.

Findings

The grading function is interpretable and musically meaningful.

It outperforms human experts in discriminating Bach chorales from generated ones.

The function effectively evaluates Transformer model outputs.

Abstract

Deep generative systems that learn probabilistic models from a corpus of existing music do not explicitly encode knowledge of a musical style, compared to traditional rule-based systems. Thus, it can be difficult to determine whether deep models generate stylistically correct output without expert evaluation, but this is expensive and time-consuming. Therefore, there is a need for automatic, interpretable, and musically-motivated evaluation measures of generated music. In this paper, we introduce a grading function that evaluates four-part chorales in the style of J.S. Bach along important musical features. We use the grading function to evaluate the output of a Transformer model, and show that the function is both interpretable and outperforms human experts at discriminating Bach chorales from model-generated ones.