构建首个主舞台电音子风格分类基准，助力精准音乐推荐与创作。
Benchmarking Sub-Genre Classification For Mainstage Dance Music
- 采用连续软标签处理混合子风格，更真实反映电音融合特性。
- 现有大模型在该任务上表现不佳，专用模型准确率显著更高。
- 数据覆盖全球顶尖电音节现场曲目，适合音乐推荐与创作研究者。
音乐分类是音乐信息检索的核心，支持多样应用。为解决主舞台电子舞曲（EDM）子风格分类缺乏全面数据集和有效方法的问题，本文提出一个新基准，包含全新数据集和基线模型。数据集扩展了子风格范畴，涵盖全球顶级DJ在大型音乐节上的最新现场演出曲目，真实呈现快速演变的电子舞曲生态。采用连续软标签方法，以捕捉多子风格混合的复杂性。实验表明，即使最先进的多模态大语言模型（MLLMs）也难以胜任此任务，而我们的专用基线模型实现高准确率。该基准可支持音乐推荐、DJ编曲和互动多媒体系统等应用，并提供视频演示。代码与数据已开源：https://github.com/Gariscat/housex-v2.git。
原文摘要 · Abstract (English)
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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。