Melody-Conditioned Lyrics Generation with SeqGANs

TL;DR

This paper introduces a novel end-to-end melody-conditioned lyrics generation system using SeqGANs, which incorporates melodic and thematic inputs to produce more meaningful lyrics without degrading performance.

Contribution

It presents the first melody-conditioned lyrics generation model based on SeqGANs that also considers thematic input, enhancing lyric relevance and quality.

Findings

Input conditions do not harm evaluation metrics.

The system generates more meaningful lyrics with additional thematic input.

The approach effectively integrates melodic and thematic information.

Abstract

Automatic lyrics generation has received attention from both music and AI communities for years. Early rule-based approaches have~---due to increases in computational power and evolution in data-driven models---~mostly been replaced with deep-learning-based systems. Many existing approaches, however, either rely heavily on prior knowledge in music and lyrics writing or oversimplify the task by largely discarding melodic information and its relationship with the text. We propose an end-to-end melody-conditioned lyrics generation system based on Sequence Generative Adversarial Networks (SeqGAN), which generates a line of lyrics given the corresponding melody as the input. Furthermore, we investigate the performance of the generator with an additional input condition: the theme or overarching topic of the lyrics to be generated. We show that the input conditions have no negative impact on the evaluation metrics while enabling the network to produce more meaningful results.