Hear: Hierarchically Enhanced Aesthetic Representations For Multidimensional Music Evaluation
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866918266001162240 |
|---|---|
| author | Liu, Shuyang Jin, Yuan Lin, Rui Chen, Shizhe Dai, Junyu Jiang, Tao |
| author_facet | Liu, Shuyang Jin, Yuan Lin, Rui Chen, Shizhe Dai, Junyu Jiang, Tao |
| contents | Evaluating song aesthetics is challenging due to the multidimensional nature of musical perception and the scarcity of labeled data. We propose HEAR, a robust music aesthetic evaluation framework that combines: (1) a multi-source multi-scale representations module to obtain complementary segment- and track-level features, (2) a hierarchical augmentation strategy to mitigate overfitting, and (3) a hybrid training objective that integrates regression and ranking losses for accurate scoring and reliable top-tier song identification. Experiments demonstrate that HEAR consistently outperforms the baseline across all metrics on both tracks of the ICASSP 2026 SongEval benchmark. The code and trained model weights are available at https://github.com/Eps-Acoustic-Revolution-Lab/EAR_HEAR. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_18869 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Hear: Hierarchically Enhanced Aesthetic Representations For Multidimensional Music Evaluation Liu, Shuyang Jin, Yuan Lin, Rui Chen, Shizhe Dai, Junyu Jiang, Tao Sound Artificial Intelligence Audio and Speech Processing Evaluating song aesthetics is challenging due to the multidimensional nature of musical perception and the scarcity of labeled data. We propose HEAR, a robust music aesthetic evaluation framework that combines: (1) a multi-source multi-scale representations module to obtain complementary segment- and track-level features, (2) a hierarchical augmentation strategy to mitigate overfitting, and (3) a hybrid training objective that integrates regression and ranking losses for accurate scoring and reliable top-tier song identification. Experiments demonstrate that HEAR consistently outperforms the baseline across all metrics on both tracks of the ICASSP 2026 SongEval benchmark. The code and trained model weights are available at https://github.com/Eps-Acoustic-Revolution-Lab/EAR_HEAR. |
| title | Hear: Hierarchically Enhanced Aesthetic Representations For Multidimensional Music Evaluation |
| topic | Sound Artificial Intelligence Audio and Speech Processing |
| url | https://arxiv.org/abs/2511.18869 |