SIGMORPHON 2023 Shared Task of Interlinear Glossing: Baseline Model

TL;DR

This paper presents a baseline transformer-based system for automating Interlinear Glossed Text generation, aiming to assist language documentation efforts by reducing manual annotation time.

Contribution

It introduces a novel approach treating gloss generation as a sequence labeling task within a shared task framework.

Findings

Baseline system achieves competitive accuracy

Transformer model effectively automates glossing process

Potential to accelerate language documentation efforts

Abstract

Language documentation is a critical aspect of language preservation, often including the creation of Interlinear Glossed Text (IGT). Creating IGT is time-consuming and tedious, and automating the process can save valuable annotator effort. This paper describes the baseline system for the SIGMORPHON 2023 Shared Task of Interlinear Glossing. In our system, we utilize a transformer architecture and treat gloss generation as a sequence labelling task.