Symbolic music generation has attracted increasing attention, while most methods focus on generating short piece (mostly less than 8 bars, and up to 32 bars). Generating long music calls for effective expression of the coherent music structure. Despite their success on long sequences, self-attention architectures still have challenge in dealing with long-term music as it requires additional care on the subtle music structure. In this paper, we propose to transfer the structure of training samples for new music generation, and develop a novel separable self-attention based model which enable the learning and transferring of the structure embedding. We show that our transfer model can generate music sequences (up to 100 bars) with interpretable structures, which bears similar structures and composition techniques with the template music from training set. Extensive experiments show its ability of generating music with target structure and well diversity. The generated 3,000 sets of music is uploaded as supplemental material.

中文翻译：

---

Melody Structure Transfer Network：用可分离的自注意力生成音乐

符号音乐的生成越来越受到关注，而大多数方法都专注于生成短片（大多小于 8 小节，最多 32 小节）。产生长音乐需要有效地表达连贯的音乐结构。尽管它们在长序列上取得了成功，但自注意力架构在处理长期音乐方面仍然存在挑战，因为它需要对微妙的音乐结构给予额外的关注。在本文中，我们建议为新的音乐生成迁移训练样本的结构，并开发一种新的基于可分离自注意力的模型，使结构嵌入的学习和迁移成为可能。我们展示了我们的传输模型可以生成具有可解释结构的音乐序列（最多 100 小节），它具有与训练集中的模板音乐相似的结构和作曲技巧。大量的实验表明它能够生成具有目标结构和多样性的音乐。生成的 3,000 组音乐作为补充材料上传。