The variational auto-encoder has become a leading framework for symbolic music generation, and a popular research direction is to study how to effectively control the generation process. A straightforward way is to control a model using different conditions during inference. However, in music practice, conditions are usually sequential (rather than simple categorical labels), involving rich information that overlaps with the learned representation. Consequently, the decoder gets confused about whether to "listen to" the latent representation or the condition, and sometimes just ignores the condition. To solve this problem, we leverage domain adversarial training to disentangle the representation from condition cues for better control. Specifically, we propose a condition corruption objective that uses the representation to denoise a corrupted condition. Minimized by a discriminator and maximized by the VAE encoder, this objective adversarially induces a condition-invariant representation. In this paper, we focus on the task of melody harmonization to illustrate our idea, while our methodology can be generalized to other controllable generative tasks. Demos and experiments show that our methodology facilitates not only condition-invariant representation learning but also higher-quality controllability compared to baselines.
翻译:变式自动编码器已成为象征性音乐生成的主导框架,而流行的研究方向是研究如何有效控制生成过程。一个直接的方法是使用不同的推论中的条件控制模型。然而,在音乐实践中,条件通常是顺序的(而不是简单的绝对标签),涉及与所学的表述方式重叠的丰富信息。因此,变式自动编码器在是否“听”潜在表达方式或条件上变得混淆不清,有时只是忽略了这个条件。为了解决这个问题,我们利用域对称培训将代表方式与条件提示分离,以更好地控制。具体地说,我们提出了一个条件腐败目标,即使用代表方式淡化一种腐蚀性的条件。这个目标被歧视者最小化,并被VAE 编码器最大化。这个目标引发了一种条件性不均匀的表达方式。在本文中,我们侧重于调和调的任务,以说明我们的想法,同时我们的方法可以推广到其他可控制的基因化任务。演示和实验表明,我们的方法不仅便于条件-变式表达方式学习,而且还有助于更高的质量控制基线。