Influence of ASR and Language Model on Alzheimer's Disease Detection

TL;DR

This paper investigates how automatic speech recognition and language models affect Alzheimer's detection accuracy from speech, highlighting the potential of acoustic features and the impact of transcription quality on diagnostic performance.

Contribution

It analyzes the influence of state-of-the-art ASR systems and language models on Alzheimer's detection, proposing a combined acoustic and lexical feature approach for improved accuracy.

Findings

Automatic transcripts without language models achieved 76.06% accuracy.

Using language models reduced accuracy by about 3%.

Acoustic features contribute significantly to detection performance.

Abstract

Alzheimer's Disease is the most common form of dementia. Automatic detection from speech could help to identify symptoms at early stages, so that preventive actions can be carried out. This research is a contribution to the ADReSSo Challenge, we analyze the usage of a SotA ASR system to transcribe participant's spoken descriptions from a picture. We analyse the loss of performance regarding the use of human transcriptions (measured using transcriptions from the 2020 ADReSS Challenge). Furthermore, we study the influence of a language model -- which tends to correct non-standard sequences of words -- with the lack of language model to decode the hypothesis from the ASR. This aims at studying the language bias and get more meaningful transcriptions based only on the acoustic information from patients. The proposed system combines acoustic -- based on prosody and voice quality -- and lexical features based on the first occurrence of the most common words. The reported results show the effect of using automatic transcripts with or without language model. The best fully automatic system achieves up to 76.06 % of accuracy (without language model), significantly higher, 3 % above, than a system employing word transcriptions decoded using general purpose language models.