arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2504.04589cs.SDeess.ASeess.SP
收藏

Solid State Bus-Comp:用于动态范围压缩器虚拟模拟建模的大规模多样化数据集

Solid State Bus-Comp: A Large-Scale and Diverse Dataset for Dynamic Range Compressor Virtual Analog Modeling

Aalto University(阿尔托大学);The Chinese University of Hong Kong, Shenzhen(香港中文大学(深圳))

Yicheng Gu, Runsong Zhang, Lauri Juvela, Zhizheng Wu

AI总结:

提出首个大规模多样化数据集Solid State Bus-Comp,用于经典VCA压缩器SSL 500 G-Bus的虚拟模拟建模,并开展基准与消融实验验证数据多样性和数量的提升效果。

摘要:

虚拟模拟(VA)建模旨在通过算法模拟硬件电路的行为,以数字方式复制其音色。动态范围压缩器(DRC)是一种音频处理模块,通过降低和放大响亮与安静声音的音量来控制音轨的动态,在音乐制作中至关重要。近年来,基于神经网络的VA建模在生成高保真模型方面展现出巨大潜力。然而,由于数据量和多样性的缺乏,它们在不同参数设置和输入声音下的泛化能力仍然有限。为了解决这个问题,我们提出了Solid State Bus-Comp,这是首个用于建模经典VCA压缩器——SSL 500 G-Bus的大规模多样化数据集。具体而言,我们从Cambridge Multitrack Library手动收集了175首未母带处理的歌曲。我们在220种参数组合下录制了压缩音频,最终形成了一个包含2528小时、涵盖多种流派、乐器、速度和调性的庞大数据集。此外,为了便于使用我们提出的数据集,我们在各种开源黑盒和灰盒模型以及白盒插件中进行了基准实验。我们还在不同数据子集中进行了消融研究,以证明提高数据多样性和数量的有效性。数据集和演示位于我们的项目页面:https://www.yichenggu.com/SolidStateBusComp/。

英文摘要:

Virtual Analog (VA) modeling aims to simulate the behavior of hardware circuits via algorithms to replicate their tone digitally. Dynamic Range Compressor (DRC) is an audio processing module that controls the dynamics of a track by reducing and amplifying the volumes of loud and quiet sounds, which is essential in music production. In recent years, neural-network-based VA modeling has shown great potential in producing high-fidelity models. However, due to the lack of data quantity and diversity, their generalization ability in different parameter settings and input sounds is still limited. To tackle this problem, we present Solid State Bus-Comp, the first large-scale and diverse dataset for modeling the classical VCA compressor -- SSL 500 G-Bus. Specifically, we manually collected 175 unmastered songs from the Cambridge Multitrack Library. We recorded the compressed audio in 220 parameter combinations, resulting in an extensive 2528-hour dataset with diverse genres, instruments, tempos, and keys. Moreover, to facilitate the use of our proposed dataset, we conducted benchmark experiments in various open-sourced black-box and grey-box models, as well as white-box plugins. We also conducted ablation studies in different data subsets to illustrate the effectiveness of the improved data diversity and quantity. The dataset and demos are on our project page: https://www.yichenggu.com/SolidStateBusComp/.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑