Raply: A profanity-mitigated rap generator

TL;DR

This paper introduces Raply, a fine-tuned GPT-2 model that generates meaningful rap lyrics with complex rhymes while reducing offensive language, marking the first effort to mitigate profanity in rap lyric generation.

Contribution

The paper presents a novel profanity-mitigated rap lyric generation model based on fine-tuning GPT-2 with a new profanity-mitigated dataset, improving both rhyme quality and content appropriateness.

Findings

Achieved high rhyme density in generated lyrics.

Successfully reduced profanity content in outputs.

First approach to profanity mitigation in rap lyric generation.

Abstract

The task of writing rap is challenging and involves producing complex rhyming schemes, yet meaningful lyrics. In this work, we propose Raply, a fine-tuned GPT-2 model capable of producing meaningful rhyming text in the style of rap. In addition to its rhyming capabilities, the model is able to generate less offensive content. It was achieved through the fine-tuning the model on a new dataset Mitislurs, a profanity-mitigated corpus. We evaluate the output of the model on two criteria: 1) rhyming based on the rhyme density metric; 2) profanity content, using the list of profanities for the English language. To our knowledge, this is the first attempt at profanity mitigation for rap lyrics generation.