用双向注意力并行生成可循环鼓点,音乐性更强。
MaskBeat: Loopable Drum Beat Generation
- 用双向注意力和迭代优化并行生成鼓点
- 生成的鼓点质量更高,音乐连贯性更好
- 适合音乐生成与自动作曲研究者
我们提出MaskBeat,一种基于Transformer的可循环鼓点生成方法。不同于逐次预测鼓击,该方法采用双向注意力结合迭代精炼,实现多乐器并行生成,同时保持音乐连贯性。此外,我们设计了针对鼓乐特性的定制损失函数,以捕捉鼓点间的音乐关系。实验表明,相比基线方法,MaskBeat生成的鼓点在质量和音乐连贯性上均有提升。
原文摘要 · Abstract (English)
We present MaskBeat, a transformer-based approach for loopable drum pattern generation. Rather than predicting drum hits sequentially, our method uses bidirectional attention with iterative refinement, allowing instruments to be generated in parallel while maintaining musical coherence. Additionally, we introduce custom loss functions that capture drum-specific musical relationships. Our experiments show that MaskBeat generates higher quality and more musically coherent drum patterns than baseline approaches.
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。