Break it Down for Me: A Study in Automated Lyric Annotation

TL;DR

This paper introduces automated lyric annotation (ALA), a new task aimed at clarifying ambiguous and jargon-filled lyrics through rephrasing and additional explanations, supported by a large annotated dataset.

Contribution

The paper defines ALA as a novel task, releases a large annotated lyric dataset, and evaluates multiple models, highlighting their unique strengths in understanding lyrics.

Findings

Models capture different types of information important for ALA

Automated and human evaluations show varied model performance

The dataset enables future research in lyric understanding

Abstract

Comprehending lyrics, as found in songs and poems, can pose a challenge to human and machine readers alike. This motivates the need for systems that can understand the ambiguity and jargon found in such creative texts, and provide commentary to aid readers in reaching the correct interpretation. We introduce the task of automated lyric annotation (ALA). Like text simplification, a goal of ALA is to rephrase the original text in a more easily understandable manner. However, in ALA the system must often include additional information to clarify niche terminology and abstract concepts. To stimulate research on this task, we release a large collection of crowdsourced annotations for song lyrics. We analyze the performance of translation and retrieval models on this task, measuring performance with both automated and human evaluation. We find that each model captures a unique type of information important to the task.