Song Aesthetics Evaluation with Multi-Stem Attention and Hierarchical Uncertainty Modeling
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912830156963840 |
|---|---|
| author | Lv, Yishan Luo, Jing Ju, Boyuan Zhang, Yang Wu, Xinda Yuan, Bo Yang, Xinyu |
| author_facet | Lv, Yishan Luo, Jing Ju, Boyuan Zhang, Yang Wu, Xinda Yuan, Bo Yang, Xinyu |
| contents | Music generative artificial intelligence (AI) is rapidly expanding music content, necessitating automated song aesthetics evaluation. However, existing studies largely focus on speech, audio or singing quality, leaving song aesthetics underexplored. Moreover, conventional approaches often predict a precise Mean Opinion Score (MOS) value directly, which struggles to capture the nuances of human perception in song aesthetics evaluation. This paper proposes a song-oriented aesthetics evaluation framework, featuring two novel modules: 1) Multi-Stem Attention Fusion (MSAF) builds bidirectional cross-attention between mixture-vocal and mixture-accompaniment pairs, fusing them to capture complex musical features; 2) Hierarchical Granularity-Aware Interval Aggregation (HiGIA) learns multi-granularity score probability distributions, aggregates them into a score interval, and applies a regression within the interval to produce the final score. We evaluated on two datasets of full-length songs: SongEval dataset (AI-generated) and an internal aesthetics dataset (human-created), and compared with two state-of-the-art (SOTA) models. Results show that the proposed method achieves stronger performance for multi-dimensional song aesthetics evaluation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_12222 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Song Aesthetics Evaluation with Multi-Stem Attention and Hierarchical Uncertainty Modeling Lv, Yishan Luo, Jing Ju, Boyuan Zhang, Yang Wu, Xinda Yuan, Bo Yang, Xinyu Sound Multimedia Audio and Speech Processing Music generative artificial intelligence (AI) is rapidly expanding music content, necessitating automated song aesthetics evaluation. However, existing studies largely focus on speech, audio or singing quality, leaving song aesthetics underexplored. Moreover, conventional approaches often predict a precise Mean Opinion Score (MOS) value directly, which struggles to capture the nuances of human perception in song aesthetics evaluation. This paper proposes a song-oriented aesthetics evaluation framework, featuring two novel modules: 1) Multi-Stem Attention Fusion (MSAF) builds bidirectional cross-attention between mixture-vocal and mixture-accompaniment pairs, fusing them to capture complex musical features; 2) Hierarchical Granularity-Aware Interval Aggregation (HiGIA) learns multi-granularity score probability distributions, aggregates them into a score interval, and applies a regression within the interval to produce the final score. We evaluated on two datasets of full-length songs: SongEval dataset (AI-generated) and an internal aesthetics dataset (human-created), and compared with two state-of-the-art (SOTA) models. Results show that the proposed method achieves stronger performance for multi-dimensional song aesthetics evaluation. |
| title | Song Aesthetics Evaluation with Multi-Stem Attention and Hierarchical Uncertainty Modeling |
| topic | Sound Multimedia Audio and Speech Processing |
| url | https://arxiv.org/abs/2601.12222 |