Guardado en:
Detalles Bibliográficos
Autores principales: Shu, Hongzhi, Li, Xinglin, Jiang, Hongyu, Fu, Minghao, Li, Xinyu
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:https://arxiv.org/abs/2409.06690
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866908477330292736
author Shu, Hongzhi
Li, Xinglin
Jiang, Hongyu
Fu, Minghao
Li, Xinyu
author_facet Shu, Hongzhi
Li, Xinglin
Jiang, Hongyu
Fu, Minghao
Li, Xinyu
contents Music classification, a cornerstone of music information retrieval, supports a wide array of applications. To address the lack of comprehensive datasets and effective methods for sub-genre classification in mainstage dance music, we introduce a novel benchmark featuring a new dataset and baseline. Our dataset expands the scope of sub-genres to reflect the diversity of recent mainstage live sets performed by leading DJs at global music festivals, capturing the vibrant and rapidly evolving electronic dance music (EDM) scene that engages millions of fans worldwide. We employ a continuous soft labeling approach to accommodate tracks blending multiple sub-genres, preserving their inherent complexity. Experiments demonstrate that even state-of-the-art multimodal large language models (MLLMs) struggle with this task, while our specialized baseline models achieve high accuracy. This benchmark supports applications such as music recommendation, DJ set curation, and interactive multimedia systems, with video demos provided. Our code and data are all open-sourced at https://github.com/Gariscat/housex-v2.git.
format Preprint
id arxiv_https___arxiv_org_abs_2409_06690
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Benchmarking Sub-Genre Classification For Mainstage Dance Music
Shu, Hongzhi
Li, Xinglin
Jiang, Hongyu
Fu, Minghao
Li, Xinyu
Sound
Artificial Intelligence
Multimedia
H.5.5; I.2.1
Music classification, a cornerstone of music information retrieval, supports a wide array of applications. To address the lack of comprehensive datasets and effective methods for sub-genre classification in mainstage dance music, we introduce a novel benchmark featuring a new dataset and baseline. Our dataset expands the scope of sub-genres to reflect the diversity of recent mainstage live sets performed by leading DJs at global music festivals, capturing the vibrant and rapidly evolving electronic dance music (EDM) scene that engages millions of fans worldwide. We employ a continuous soft labeling approach to accommodate tracks blending multiple sub-genres, preserving their inherent complexity. Experiments demonstrate that even state-of-the-art multimodal large language models (MLLMs) struggle with this task, while our specialized baseline models achieve high accuracy. This benchmark supports applications such as music recommendation, DJ set curation, and interactive multimedia systems, with video demos provided. Our code and data are all open-sourced at https://github.com/Gariscat/housex-v2.git.
title Benchmarking Sub-Genre Classification For Mainstage Dance Music
topic Sound
Artificial Intelligence
Multimedia
H.5.5; I.2.1
url https://arxiv.org/abs/2409.06690