SDAIMMASDec 10, 2024

Frechet Music Distance: A Metric For Generative Symbolic Music Evaluation

arXiv:2412.07948v212 citationsh-index: 2
Originality Incremental advance
AI Analysis

This provides a domain-specific evaluation metric for researchers in symbolic music generation, establishing a reproducible standard.

The paper tackles the problem of evaluating generative symbolic music models by introducing the Frechet Music Distance (FMD), a metric that calculates distances between distributions of music embeddings, and results show it effectively differentiates model quality across datasets.

In this paper we introduce the Frechet Music Distance (FMD), a novel evaluation metric for generative symbolic music models, inspired by the Frechet Inception Distance (FID) in computer vision and Frechet Audio Distance (FAD) in generative audio. FMD calculates the distance between distributions of reference and generated symbolic music embeddings, capturing abstract musical features. We validate FMD across several datasets and models. Results indicate that FMD effectively differentiates model quality, providing a domain-specific metric for evaluating symbolic music generation, and establishing a reproducible standard for future research in symbolic music modeling.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes