arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2404.09313eess.AScs.AI
收藏

Text-to-Song:面向融合人声与伴奏的可控音乐生成

Text-to-Song: Towards Controllable Music Generation Incorporating Vocals and Accompaniment

Zhejiang University(浙江大学)

Zhiqing Hong, Rongjie Huang, Xize Cheng, Yongqi Wang, Ruiqi Li, Fuming You, Zhou Zhao, Zhimeng Zhang

AI总结:

针对现有研究未兼顾歌声与伴奏合成的问题,本文提出文本到歌曲合成新任务,开发两阶段方法Melodist,通过三塔对比预训练实现可控生成,并构建中文歌曲数据集验证了其合成质量与风格一致性。

摘要:

歌曲是歌声与伴奏的结合体。然而现有研究分别聚焦于歌声合成与音乐生成,对歌曲合成的探索关注较少。本文提出一项名为text-to-song synthesis(文本到歌曲合成)的新任务,同时涵盖人声与伴奏生成。我们开发了Melodist,一种两阶段文本到歌曲方法,由singing voice synthesis (SVS,歌声合成)与vocal-to-accompaniment (V2A,人声转伴奏)合成组成。Melodist利用三塔对比预训练学习更有效的文本表示,以实现可控的V2A合成。为缓解研究中的数据稀缺问题,我们构建了一个从音乐网站挖掘的中文歌曲数据集。在该数据集上的评估结果表明,Melodist合成的歌曲在质量与风格一致性上表现可比。音频样本可参见https://text2songMelodist.github.io/Sample/。

英文摘要:

A song is a combination of singing voice and accompaniment. However, existing works focus on singing voice synthesis and music generation independently. Little attention was paid to explore song synthesis. In this work, we propose a novel task called text-to-song synthesis which incorporating both vocals and accompaniments generation. We develop Melodist, a two-stage text-to-song method that consists of singing voice synthesis (SVS) and vocal-to-accompaniment (V2A) synthesis. Melodist leverages tri-tower contrastive pretraining to learn more effective text representation for controllable V2A synthesis. A Chinese song dataset mined from a music website is built up to alleviate data scarcity for our research. The evaluation results on our dataset demonstrate that Melodist can synthesize songs with comparable quality and style consistency. Audio samples can be found in https://text2songMelodist.github.io/Sample/.

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。

↑