# Impact of ASR on Alzheimer's Disease Detection: All Errors are Equal, but Deletions are More Equal than Others

## TL;DR

This study examines how different types of ASR errors impact Alzheimer's detection, revealing that deletion errors significantly hinder performance and suggesting ASR optimization to penalize deletions could enhance detection accuracy.

## Contribution

The paper demonstrates the differential impact of ASR error types on dementia detection and proposes targeted ASR improvements to boost model performance.

## Key findings

- Deletion errors most negatively affect detection accuracy
- The impact of error types is consistent across datasets
- Optimizing ASR to penalize deletions improves dementia detection

## Abstract

Automatic Speech Recognition (ASR) is a critical component of any fully-automated speech-based dementia detection model. However, despite years of speech recognition research, little is known about the impact of ASR accuracy on dementia detection. In this paper, we experiment with controlled amounts of artificially generated ASR errors and investigate their influence on dementia detection. We find that deletion errors affect detection performance the most, due to their impact on the features of syntactic complexity and discourse representation in speech. We show the trend to be generalisable across two different datasets for cognitive impairment detection. As a conclusion, we propose optimising the ASR to reflect a higher penalty for deletion errors in order to improve dementia detection performance.