Can Machines Generate Personalized Music? A Hybrid Favorite-aware Method for User Preference Music Transfer

TL;DR

This paper explores the novel problem of user preference music transfer (UPMT), aiming to generate personalized music by transferring user preferences, and introduces a hybrid favorite-aware method to address this challenge.

Contribution

It proposes a new hybrid favorite-aware approach for personalized music transfer, filling a gap in existing music style transfer research.

Findings

Demonstrates effectiveness of the proposed method in personalized music transfer

Achieves higher user satisfaction compared to baseline models

Provides a new framework for user preference modeling in music

Abstract

User preference music transfer (UPMT) is a new problem in music style transfer that can be applied to many scenarios but remains understudied.