In BLOOM: Creativity and Affinity in Artificial Lyrics and Art

TL;DR

This paper explores the use of BLOOM-176B for generating Chinese song lyrics and album art, highlighting the limitations of current evaluation metrics and introducing a new dataset for future research.

Contribution

It demonstrates the application of a large multilingual language model for creative lyric generation and multimodal art creation, and introduces the MojimLyrics dataset for Chinese lyrics.

Findings

Human evaluation shows limitations of MAUVE in assessing creativity.

Humans rate some machine-generated lyrics higher than real lyrics.

A stable diffusion model produces high-quality lyric-guided album art.

Abstract

We apply a large multilingual language model (BLOOM-176B) in open-ended generation of Chinese song lyrics, and evaluate the resulting lyrics for coherence and creativity using human reviewers. We find that current computational metrics for evaluating large language model outputs (MAUVE) have limitations in evaluation of creative writing. We note that the human concept of creativity requires lyrics to be both comprehensible and distinctive -- and that humans assess certain types of machine-generated lyrics to score more highly than real lyrics by popular artists. Inspired by the inherently multimodal nature of album releases, we leverage a Chinese-language stable diffusion model to produce high-quality lyric-guided album art, demonstrating a creative approach for an artist seeking inspiration for an album or single. Finally, we introduce the MojimLyrics dataset, a Chinese-language dataset of popular song lyrics for future research.