Infusing Acoustic Pause Context into Text-Based Dementia Assessment

TL;DR

This paper explores how incorporating acoustic pause information into transformer models improves the detection of cognitive impairment levels in dementia patients using speech analysis.

Contribution

It introduces a novel approach of integrating pause-enriched transcripts into transformer models for dementia assessment, highlighting the importance of context-specific features.

Findings

Pause information enhances model accuracy

Different tasks benefit from different speech features

Acoustic cross-attention improves classification performance

Abstract

Speech pauses, alongside content and structure, offer a valuable and non-invasive biomarker for detecting dementia. This work investigates the use of pause-enriched transcripts in transformer-based language models to differentiate the cognitive states of subjects with no cognitive impairment, mild cognitive impairment, and Alzheimer's dementia based on their speech from a clinical assessment. We address three binary classification tasks: Onset, monitoring, and dementia exclusion. The performance is evaluated through experiments on a German Verbal Fluency Test and a Picture Description Test, comparing the model's effectiveness across different speech production contexts. Starting from a textual baseline, we investigate the effect of incorporation of pause information and acoustic context. We show the test should be chosen depending on the task, and similarly, lexical pause information and acoustic cross-attention contribute differently.