Augmenting Poetry Composition with Verse by Verse

TL;DR

This paper introduces Verse by Verse, an AI system that assists users in writing poetry by suggesting lines inspired by classic American poets, combining generative and recommendation models.

Contribution

The paper presents a novel AI poetry assistant that integrates generative and retrieval models to support creative poetry composition.

Findings

AI poets styled after American classics generate diverse verse suggestions.

The system effectively recommends next lines based on previous verse.

The approach enhances creative writing with AI-generated poetic suggestions.

Abstract

We describe Verse by Verse, our experiment in augmenting the creative process of writing poetry with an AI. We have created a group of AI poets, styled after various American classic poets, that are able to offer as suggestions generated lines of verse while a user is composing a poem. In this paper, we describe the underlying system to offer these suggestions. This includes a generative model, which is tasked with generating a large corpus of lines of verse offline and which are then stored in an index, and a dual-encoder model that is tasked with recommending the next possible set of verses from our index given the previous line of verse.