Generating Completions for Broca's Aphasic Sentences Using Large Language Models

TL;DR

This study investigates the use of large language models to complete sentences characteristic of Broca's aphasia, demonstrating their potential to improve communication aids for affected individuals.

Contribution

The paper introduces a method to fine-tune LLMs using synthetic aphasic data for sentence completion in Broca's aphasia, without relying on authentic samples.

Findings

LLMs can reconstruct agrammatic sentences effectively.

Performance improves with longer input utterances.

Synthetic data can be used to train models for clinical language tasks.

Abstract

Broca's aphasia is a type of aphasia characterized by non-fluent, effortful and agrammatic speech production with relatively good comprehension. Since traditional aphasia treatment methods are often time-consuming, labour-intensive, and do not reflect real-world conversations, applying natural language processing based approaches such as Large Language Models (LLMs) could potentially contribute to improving existing treatment approaches. To address this issue, we explore the use of sequence-to-sequence LLMs for completing Broca's aphasic sentences. We first generate synthetic Broca's aphasic data using a rule-based system designed to mirror the linguistic characteristics of Broca's aphasic speech. Using this synthetic data (without authentic aphasic samples), we then fine-tune four pre-trained LLMs on the task of completing agrammatic sentences. We evaluate our fine-tuned models on both synthetic and authentic Broca's aphasic data. We demonstrate LLMs' capability for reconstructing agrammatic sentences, with the models showing improved performance with longer input utterances. Our result highlights the LLMs' potential in advancing communication aids for individuals with Broca's aphasia and possibly other clinical populations.