Computational Music

Computational music research aims to understand and generate music using computational methods, focusing on tasks like music transcription, analysis, and generation. Current research heavily utilizes deep learning models, particularly transformers and coupled hidden Markov models, to analyze musical structure, extract features (e.g., melody, harmony), and generate new musical pieces in various styles. This field contributes to both music theory and practice by providing objective analyses of musical works, creating new tools for composers and performers, and offering insights into the relationship between music and other aspects of human culture and behavior.