Combinatorial music generation model with song structure graph analysis
This is an incremental improvement for symbolic music generation, potentially useful in Music Information Retrieval applications like composition and classification.
The researchers tackled symbolic music generation by proposing a model that constructs a song structure graph with note sequences and instruments as node features and correlations as edge features, then uses a Graph Neural Network and Unet to generate pianoroll images; the result shows the model can generate comprehensive music forms.
In this work, we propose a symbolic music generation model with the song structure graph analysis network. We construct a graph that uses information such as note sequence and instrument as node features, while the correlation between note sequences acts as the edge feature. We trained a Graph Neural Network to obtain node representation in the graph, then we use node representation as input of Unet to generate CONLON pianoroll image latent. The outcomes of our experimental results show that the proposed model can generate a comprehensive form of music. Our approach represents a promising and innovative method for symbolic music generation and holds potential applications in various fields in Music Information Retreival, including music composition, music classification, and music inpainting systems.