WuYun: Exploring hierarchical skeleton-guided melody generation using knowledge-enhanced deep learning
This work addresses the challenge of improving melody structure in music generation for applications in creative AI, offering a hybrid approach that combines data-driven and knowledge-based methods.
The paper tackles the problem of generating structured melodies by introducing WuYun, a knowledge-enhanced deep learning architecture that first creates a melodic skeleton and then infills it, resulting in melodies with better long-term structure and musicality, outperforming state-of-the-art methods by an average of 0.51 on subjective metrics.
Although deep learning has revolutionized music generation, existing methods for structured melody generation follow an end-to-end left-to-right note-by-note generative paradigm and treat each note equally. Here, we present WuYun, a knowledge-enhanced deep learning architecture for improving the structure of generated melodies, which first generates the most structurally important notes to construct a melodic skeleton and subsequently infills it with dynamically decorative notes into a full-fledged melody. Specifically, we use music domain knowledge to extract melodic skeletons and employ sequence learning to reconstruct them, which serve as additional knowledge to provide auxiliary guidance for the melody generation process. We demonstrate that WuYun can generate melodies with better long-term structure and musicality and outperforms other state-of-the-art methods by 0.51 on average on all subjective evaluation metrics. Our study provides a multidisciplinary lens to design melodic hierarchical structures and bridge the gap between data-driven and knowledge-based approaches for numerous music generation tasks.