"SongEval: A Benchmark Dataset for Song Aesthetics Evaluation" by Huixin Xue

  • Published on 23 déc. 2025 by xuexue1

Aesthetics serve as an implicit and important criterion in song generation tasks that reflect human perception beyond objective metrics. However, evaluating the aesthetics of generated songs remains a fundamental challenge, as the appreciation of music is highly subjective. Existing evaluation metrics, such as embedding-based distances, are limited in reflecting the subjective and perceptual aspects that define musical appeal. To address this issue, we introduce SongEval, the first open-source, large-scale benchmark dataset for evaluating the aesthetics of full-length songs.

➡️ This presentation is part of IRCAM Forum Workshops Paris / Enghien-les-Bains March 2026

SongEval includes over 2,399 songs in full length, summing up to more than 140 hours, with aesthetic ratings from 16 professional annotators with musical backgrounds. Each song is evaluated across five key dimensions: overall coherence, memorability, naturalness of vocal breathing and phrasing, clarity of song structure, and overall musicality. The dataset covers both English and Chinese songs, spanning nine mainstream genres. Moreover, to assess the effectiveness of song aesthetic evaluation, we conduct experiments using SongEval to predict aesthetic scores and demonstrate better performance than existing objective evaluation metrics in predicting human-perceived musical quality, We provide the dataset and toolkit for song aesthetic evaluation at https://huggingface.co/datasets/ASLP-lab/SongEval and https://github.com/ASLP-lab/SongEval