Perceiving Music Quality with GANs

TL;DR

This paper introduces a no-reference music quality assessment method using a GAN discriminator trained on a large music library, which correlates well with human subjective ratings and does not require access to original unaltered signals.

Contribution

The paper proposes a novel unsupervised approach leveraging GAN discriminators for no-reference music quality assessment, enabling applications without needing paired reference signals.

Findings

Discriminator scores significantly correlate with human ratings.

Method works across various types of signal degradations.

No-reference assessment is feasible without access to original signals.

Abstract

Several methods have been developed to assess the perceptual quality of audio under transforms like lossy compression. However, they require paired reference signals of the unaltered content, limiting their use in applications where references are unavailable. This has hindered progress in audio generation and style transfer, where a no-reference quality assessment method would allow more reproducible comparisons across methods. We propose training a GAN on a large music library, and using its discriminator as a no-reference quality assessment measure of the perceived quality of music. This method is unsupervised, needs no access to degraded material and can be tuned for various domains of music. In a listening test with 448 human subjects, where participants rated professionally produced music tracks degraded with different levels and types of signal degradations such as waveshaping distortion and low-pass filtering, we establish a dataset of human rated material. By using the human rated dataset we show that the discriminator score correlates significantly with the subjective ratings, suggesting that the proposed method can be used to create a no-reference musical audio quality assessment measure.