arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2409.20196cs.SDcs.AIeess.AS
收藏

旋律引导的音乐生成

Melody-Guided Music Generation

School of Business, Guangxi University(广西大学商学院);School of Computing and Artificial Intelligence, Southwestern University of Finance and Economics(西南财经大学计算机与人工智能学院);School of Business Administration, Southwestern University of Finance and Economics(西南财经大学工商管理学院)

Shaopeng Wei, Manzhen Wei, Haoyu Wang, Yu Zhao, Gang Kou

AI总结:

本文提出旋律引导的音乐生成模型MG2,通过对比语言-音乐预训练和检索增强扩散,在极少参数和训练数据下超越现有开源文本到音乐生成模型。

摘要:

我们提出旋律引导的音乐生成(MG2)模型,这是一种利用旋律来引导文本到音乐生成的新方法,尽管方法简单且资源有限,却取得了优异的性能。具体来说,我们首先使用新提出的对比语言-音乐预训练(Contrastive Language-Music Pretraining)将文本与音频波形及其关联旋律对齐,使学习到的文本表示融合了隐式旋律信息。随后,我们在文本提示和检索到的旋律两者上对检索增强扩散模块进行条件化。这使得MG2能够生成既反映给定文本描述内容,又在显式旋律信息引导下保持内在和谐性的音乐。我们在两个公开数据集MusicCaps和MusicBench上进行了大量实验。令人惊讶的是,实验结果表明,所提出的MG2模型超越了当前的开源文本到音乐生成模型,并且与最先进的同类模型相比,其参数量不到1/3,或训练数据不到1/200。此外,我们进行了全面的人工评估,涉及三类用户和五个视角,使用新设计的问卷来探索MG2在现实世界中的潜在应用。

英文摘要:

We present the Melody-Guided Music Generation (MG2) model, a novel approach using melody to guide the text-to-music generation that, despite a simple method and limited resources, achieves excellent performance. Specifically, we first align the text with audio waveforms and their associated melodies using the newly proposed Contrastive Language-Music Pretraining, enabling the learned text representation fused with implicit melody information. Subsequently, we condition the retrieval-augmented diffusion module on both text prompt and retrieved melody. This allows MG2 to generate music that reflects the content of the given text description, meantime keeping the intrinsic harmony under the guidance of explicit melody information. We conducted extensive experiments on two public datasets: MusicCaps and MusicBench. Surprisingly, the experimental results demonstrate that the proposed MG2 model surpasses current open-source text-to-music generation models, achieving this with fewer than 1/3 of the parameters or less than 1/200 of the training data compared to state-of-the-art counterparts. Furthermore, we conducted comprehensive human evaluations involving three types of users and five perspectives, using newly designed questionnaires to explore the potential real-world applications of MG2.

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。

↑