Music Sentiment Transfer
This work addresses sentiment transfer in music, a domain-specific problem with incremental contributions from style transfer methods.
The paper tackles the novel task of music sentiment transfer, applying sentiment from a source to a target piece of music using CycleGAN with symbolic MIDI data, achieving results that preserve content and realism but indicate greater difficulty compared to image sentiment transfer due to music's temporal nature and dataset scarcity.
Music sentiment transfer is a completely novel task. Sentiment transfer is a natural evolution of the heavily-studied style transfer task, as sentiment transfer is rooted in applying the sentiment of a source to be the new sentiment for a target piece of media; yet compared to style transfer, sentiment transfer has been only scantily studied on images. Music sentiment transfer attempts to apply the high level objective of sentiment transfer to the domain of music. We propose CycleGAN to bridge disparate domains. In order to use the network, we choose to use symbolic, MIDI, data as the music format. Through the use of a cycle consistency loss, we are able to create one-to-one mappings that preserve the content and realism of the source data. Results and literature suggest that the task of music sentiment transfer is more difficult than image sentiment transfer because of the temporal characteristics of music and lack of existing datasets.