Explainable Detection of Machine Generated Music and Early Systematic Evaluation

TL;DR

This paper systematically evaluates machine-generated music detection models, compares their performance, and explores explainability techniques to understand their decision-making, aiming to improve MGM detection methods.

Contribution

It provides the first comprehensive systematic evaluation of MGM detection models across various architectures and introduces explainability analysis to interpret model decisions.

Findings

ResNet18 outperforms other models in detection accuracy.

Multimodal models are effective due to music's inherent multimodal nature.

Explainability tools reveal key features influencing model decisions.

Abstract

Machine-generated music (MGM) has become a groundbreaking innovation with wide-ranging applications, such as music therapy, personalised editing, and creative inspiration within the music industry. However, the unregulated proliferation of MGM presents considerable challenges to the entertainment, education, and arts sectors by potentially undermining the value of high-quality human compositions. Consequently, MGM detection (MGMD) is crucial for preserving the integrity of these fields. Despite its significance, MGMD domain lacks comprehensive systematic evaluation results necessary to drive meaningful progress. To address this gap, we conduct experiments on existing large-scale datasets using a range of foundational models for audio processing, establishing systematic evaluation results tailored to the MGMD task. Our selection includes traditional machine learning models, deep neural networks, Transformer-based architectures, and State space models (SSM). Recognising the inherently multimodal nature of music, which integrates both melody and lyrics, we also explore fundamental multimodal models in our experiments. Beyond providing basic binary classification outcomes, we delve deeper into model behaviour using multiple explainable Artificial Intelligence (XAI) tools, offering insights into their decision-making processes. Our analysis reveals that ResNet18 performs the best according to in-domain and out-of-domain tests. By providing a comprehensive comparison of systematic evaluation results and their interpretability, we propose several directions to inspire future research to develop more robust and effective detection methods for MGM. We provide our codes and some samples on Github repository.