Evaluating the Evaluation of Music Generation Models with Generative Psychometrics
Organizations: Departamento de Ciência da Computação, UFMG – Belo Horizonte, MG, Brazil · Escola de Música, UFMG – Belo Horizonte, MG, Brazil
Abstract
Subjective user studies are one of the major methods for evaluating music generation models. However, subjective evaluation protocols vary widely and are rarely specified in enough detail to be reproduced, hindering comparison across studies. In this paper, we map subjective evaluation methods in ISMIR music generation papers from 2020-2025 and apply generative psychometrics to assess the stability of their items. We extract 284 items from the papers' methodological descriptions, using an LLM and manual validation to standardize them. We represent these items as contextual embeddings and apply Unique Variable Analysis to remove redundant, locally dependent items and Bootstrap Exploratory Graph Analysis to estimate their latent dimensions. Results show the network converges onto a taxonomy of six stable quality dimensions: Rhythm, Melody, Audio, Composition, Structure, and Lyrics, pointing to a unified methodology for conducting user studies to evaluate music generation models.
Figures & tables
| Dimension | Most Representative Item | NL |
|---|---|---|
| Rhythm | “Rhythmic stability (whether the drummer follows a steady tempo) of the generated drum.” | 0.59 |
| Melody | “What is the overall quality of the melody?” | 0.55 |
| Audio | “A subjective assessment of the audio quality.” | 0.53 |
| Composition | “I like this composition.” | 0.52 |
| Structure | “Long-term structure.” | 0.61 |
| Lyrics | “Are the impressions of the input text and the generated lyrics similar?” | 0.50 |
Music Generation Evaluation
Music Generation Evaluation measures the quality, musical structure, prompt alignment, controllability, and robustness of generated music.
See all 43 papers