Towards Assessing Data Replication in Music Generation with Music Similarity Metrics on Raw Audio

TL;DR

This paper introduces the MiRA tool, an open evaluation method using music similarity metrics to assess data replication in AI-generated music, addressing ethical and legal concerns of copyright infringement.

Contribution

The paper presents a novel, model-independent evaluation tool for detecting data replication in music generation, validated through experiments across genres with synthetic samples.

Findings

The methodology can estimate data replication above 10%.

Five similarity metrics effectively identify exact replication.

The tool promotes ethical evaluation of music generative models.

Abstract

Recent advancements in music generation are raising multiple concerns about the implications of AI in creative music processes, current business models and impacts related to intellectual property management. A relevant discussion and related technical challenge is the potential replication and plagiarism of the training set in AI-generated music, which could lead to misuse of data and intellectual property rights violations. To tackle this issue, we present the Music Replication Assessment (MiRA) tool: a model-independent open evaluation method based on diverse audio music similarity metrics to assess data replication. We evaluate the ability of five metrics to identify exact replication by conducting a controlled replication experiment in different music genres using synthetic samples. Our results show that the proposed methodology can estimate exact data replication with a proportion higher than 10%. By introducing the MiRA tool, we intend to encourage the open evaluation of music-generative models by researchers, developers, and users concerning data replication, highlighting the importance of the ethical, social, legal, and economic consequences. Code and examples are available for reproducibility purposes.