Let the Poem Hit the Rhythm: Using a Byte-Based Transformer for Beat-Aligned Poetry Generation

TL;DR

This paper presents a byte-based transformer model that generates poetry aligned with musical beat patterns, demonstrating promising beat synchronization and semantic coherence in computational creativity.

Contribution

It introduces a novel application of a byte-based transformer for beat-aligned poetry generation, bridging poetry and music through computational methods.

Findings

High beat alignment in generated poems

Maintains semantic coherence in outputs

Demonstrates potential for computational creativity

Abstract

The intersection between poetry and music provides an interesting case for computational creativity, yet remains relatively unexplored. This paper explores the integration of poetry and music through the lens of beat patterns, investigating whether a byte-based language model can generate words that fit specific beat patterns within the context of poetry. Drawing on earlier studies, we developed a method to train a byte-based transformer model, ByT5, to align poems with beat patterns. The results demonstrate a high level of beat alignment while maintaining semantic coherence. Future work will aim to improve the model's ability to create complete beat-aligned poems.