q-bio.QMMar 18, 2026

Impact of automatic speech recognition quality on Alzheimer's disease detection from spontaneous speech: a reproducible benchmark study with lexical modeling and statistical validation

Authors: Himadri S Samanta

Organizations: Independent Researcher, Austin, Texas, USA

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

Early detection of Alzheimer's disease from spontaneous speech has emerged as a promising non-invasive screening approach. However, the influence of automatic speech recognition (ASR) quality on downstream clinical language modeling remains insufficiently understood. In this study, we investigate Alzheimer's disease detection using lexical features derived from Whisper ASR transcripts on the ADReSSo 2021 diagnosis dataset. We evaluate interpretable machine-learning models, including Logistic Regression and Linear Support Vector Machines, using TF-IDF text representations under repeated 5x5 stratified cross-validation. Our results demonstrate that transcript quality has a statistically significant impact on classification performance. Models trained on Whisper-small transcripts consistently outperform those using Whisper-base transcripts, achieving balanced accuracy above 0.7850 with Linear SVM. Paired statistical testing confirms that the observed improvements are significant. Importantly, classifier complexity contributes less to performance variation than ASR transcription quality. Feature analysis reveals that cognitively normal speakers produce more semantically precise object- and scene-descriptive language, whereas Alzheimer's speech is characterized by vagueness, discourse markers, and increased hesitation patterns. These findings suggest that high-quality ASR can enable simple, interpretable lexical models to achieve competitive Alzheimer's detection performance without explicit acoustic modeling. The study provides a reproducible benchmark pipeline and highlights ASR selection as a critical modeling decision in clinical speech-based artificial intelligence systems.

Figures & tables

SplitADCN
Training8779
Official test-dist3536
Table 1: ADReSSo diagnosis dataset configuration used in this study.
ClassifierASR modelMean balanced accuracyStandard deviation
Logistic RegressionWhisper base0.74910.0814
Logistic RegressionWhisper small0.77570.0703
Linear SVMWhisper base0.73540.0865
Linear SVMWhisper small0.78500.0745
Table 2: Repeated stratified 5 ×\times 5 cross-validation results on the ADReSSo training split.
ComparisonMean differencett statisticTwo-sided ppInterpretation
Whisper small vs base (LR)+0.02662.44900.0220significant
Whisper small vs base (SVM)+0.04973.88320.0007highly significant
Linear SVM vs LR (Whisper small)+0.00931.05510.3019not significant
Table 3: Paired statistical comparisons over 25 repeated-cross-validation folds.
Figure 1: Speech-based Alzheimer’s disease detection pipeline used in this work. Audio recordings are transcribed with Whisper ASR, transformed into TF–IDF features, classified with linear models, and evaluated using repeated cross-validation and statistical testing.
Figure 2: Repeated 5 ×\times 5 cross-validation balanced accuracy for each classifier/ASR combination. Error bars denote standard deviation across 25 folds.
Figure 3: Effect sizes (Cohen’s dd ) for key paired comparisons. The ASR effect is larger than the classifier effect.
Figure 4: Out-of-fold ROC curve for logistic regression trained on Whisper-small transcripts. AUC = 0.8532.
Figure 5: Aggregate confusion matrix (raw counts) for logistic regression with Whisper-small transcripts across repeated folds.
Figure 6: Normalized confusion matrix for logistic regression with Whisper-small transcripts. CN recall = 0.742; AD recall = 0.789.
AD-indicative termCoefficientCN-indicative termCoefficient
going+0.1718open-0.1295
okay+0.1540window-0.1275
happening+0.1444reaching-0.1196
tell+0.1212overflowing-0.0998
picture+0.1132sink-0.0978
going picture+0.1114cookie-0.0912
Table 4: Selected high-magnitude lexical indicators from the logistic-regression model trained on Whisper-small transcripts. Positive coefficients increase the probability of AD; negative coefficients increase the probability of cognitively normal speech.
Figure 7: Top AD-indicative lexical terms from the logistic-regression model trained on Whisper-small transcripts.
Figure 8: Top CN-indicative lexical terms from the logistic-regression model trained on Whisper-small transcripts.
More in this topic

Speech-Based Dementia Detection

Speech-based dementia detection uses spoken language and vocal patterns to identify cognitive impairment and support accessible screening.

See all 44 papers

Explore similar work

CardsList
Jul 11, 2026

Transcript-Free Lightweight Detection of Alzheimer's Disease from Spontaneous Speech Using Handcrafted MFCC-Dominant Acoustic Biomarkers

It is still hard to find Alzheimer's disease (AD) early, especially when neuroimaging is expensive or tools that depend on language are not available. Spontaneous speech provides a non-invasive signal; however, numerous current methodologies depend on transcripts/ASR or computationally intensive deep models. We offer a simple, audio-only baseline for detecting AD using 176 Cookie Theft recordings from the DementiaBank Pitt corpus (88 AD, 88 controls). WebRTC voice activity detection (VAD) is used to separate speech from non-speech. We take out 99 hand-crafted acoustic-temporal features, including pause and fluency statistics, spectral/prosodic descriptors, and MFCC summaries with Δ and ΔΔ. Evaluation is performed using a stringent speaker-independent GroupShuffleSplit,documenting performance across 30 iterations. A lightweight SVM with an RBF kernel gets an average AUC of 0.674 across runs. For example, a single split has an AUC of 0.742 and an accuracy of 0.657. We also present an exploratory compact-feature analysis utilizing a Top-20 subset ranked by Random Forest importance; since selection is not nested within training splits, these results may be overly optimistic and are not employed for primary conclusions (AUC 0.719). The results indicate that transcript-free spectro-temporal and fluency-related cues can facilitate speaker-independent Alzheimer's disease screening from raw audio, establishing a practical foundation for deployment-oriented research.

Rashin Gholijani Farahani, Azam Bastanfard

Department of computer Engineering, Ka.c., Islamic Azad University Karaj,Alborz, Iran

Speech-Based Dementia DetectionSpeech ProcessingAlzheimer's Disease Classification
AskPDF
Aug 7, 2026

LSEAD: A Privacy-Preserving LLM-Based Speech Analysis Framework for Early Alzheimer's Disease Screening

Early diagnosis of Alzheimer's disease (AD) is critical for enabling timely interventions that may slow disease progression and improve patient outcomes. There is a growing need for AD detection methods that are non-invasive and cost-effective, especially in real-world clinical settings with diverse patient populations and recording conditions. Speech-based screening addresses these needs by using natural speech collected without specialized equipment. Recent advances in large language models (LLMs) have improved speech analysis by providing rich linguistic representations and strong generalization. In this study, we propose LSEAD, a speech-based AD detection framework using pretrained open-source LLMs. Speech recordings are automatically transcribed, and text embeddings are extracted using locally deployed LLMs. Principal component analysis (PCA) is applied to reduce dimensionality before classification. Because the framework relies only on speech transcripts and locally deployed models, it supports privacy-preserving AD risk assessment without external data exchange. We evaluate LSEAD on the ADReSS20 and ADReSSo2021 benchmark datasets. Experimental results show that LLM-based embeddings generalize well across datasets and improve AD classification accuracy by up to 5 percent over existing methods, especially for early-stage detection. These results demonstrate that LSEAD provides a practical, secure, and scalable approach for early AD screening.

Xin Wang, Yingchao Huang, Yuhan Su +2

Faculty of Digital Innovation, Arts & Sciences, Saskatchewan Polytechnic, Regina SK S4S 5X1, Canada · School of Basic Medical Sciences, Hebei University, Baoding 071000, China · Department of Civil & Environmental Engineering and School of Mining & Petroleum Engineering, University of Alberta, Edmonton AB T6G 2H5, Canada +1

Speech-Based Dementia DetectionAlzheimer's Disease Classification
AskPDF
Aug 31, 2026

Cleaner Speech, Weaker Generalization: Revisiting Pitt-Derived Benchmarks for Alzheimer's Disease Detection

Speech-based Alzheimer's disease (AD) detection increasingly relies on speech-enhanced and curated versions of the Pitt Corpus, where speech enhancement, sample selection, and demographic balancing are often treated as beneficial preprocessing steps. However, whether these transformations improve real-world AD detection or instead affect model generalization and prediction behavior remains unclear. In this work, we revisit the role of speech preprocessing and dataset curation across widely used benchmarks for speech-based AD detection. We evaluate the speech quality of different datasets, the cross-dataset generalization of multiple deep learning models under matched and mismatched enhancement settings, and the behavior of several recent large audio-language models (LALMs). Experimental results show that across multiple supervised speech models, speech-enhanced datasets often improve in-domain performance while reducing robustness in cross-domain evaluation. Matched enhancement between training and test data alleviates, but does not eliminate, this degradation. LALMs show a similar sensitivity: enhanced datasets induce stronger class imbalance and prediction shifts than unprocessed data. These results suggest that speech preprocessing and dataset curation can substantially influence downstream AD detection behavior, indicating that ``cleaner'' speech datasets are not necessarily more reliable for real-world AD detection.

Luqi Sun, Shreeram Suresh Chandra, Lin Zhang +5

Center for Language and Speech Processing (CLSP), Johns Hopkins University · Research Center for Information Technology Innovation, Academia Sinica · University of Michigan, Ann Arbor +1

Speech-Based Dementia DetectionSpeech EnhancementAlzheimer's Disease ClassificationCross-Dataset Generalization
AskPDF