Beyond Binary: Multiclass Paraphasia Detection with Generative Pretrained Transformers and End-to-End Models

TL;DR

This paper introduces novel end-to-end models, including a GPT-based approach, for multiclass paraphasia detection in speech transcripts, advancing beyond binary detection methods.

Contribution

It presents new end-to-end and GPT-based models specifically designed for multiclass paraphasia detection, a previously unexplored area.

Findings

Single sequence model outperforms GPT baseline

End-to-end models effectively identify multiple paraphasia types

Approaches integrate speech recognition and classification

Abstract

Aphasia is a language disorder that can lead to speech errors known as paraphasias, which involve the misuse, substitution, or invention of words. Automatic paraphasia detection can help those with Aphasia by facilitating clinical assessment and treatment planning options. However, most automatic paraphasia detection works have focused solely on binary detection, which involves recognizing only the presence or absence of a paraphasia. Multiclass paraphasia detection represents an unexplored area of research that focuses on identifying multiple types of paraphasias and where they occur in a given speech segment. We present novel approaches that use a generative pretrained transformer (GPT) to identify paraphasias from transcripts as well as two end-to-end approaches that focus on modeling both automatic speech recognition (ASR) and paraphasia classification as multiple sequences vs. a single sequence. We demonstrate that a single sequence model outperforms GPT baselines for multiclass paraphasia detection.