[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-85831-en":3,"doc-seo-85831-105":29,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":13,"seo_description":14,"update_tm":27,"read_time":28},85831,8796095461610,"Oliver","https://ap-avatar.wpscdn.com/davatar_276721f389ce27ea32af1340a28f341c",8,"Research & Report","Transcript-Free Lightweight Detection of Alzheimer's Disease from Spontaneous Speech Using Handcrafted MFCC-Dominant Acoustic Biomarkers","Early Alzheimer’s disease detection is challenging due to expensive neuroimaging and the absence of transcript-dependent language tools. This work leverages spontaneous speech as a non-invasive signal and proposes a lightweight, audio-only, transcript-free baseline for screening using 176 Cookie Theft recordings from DementiaBank Pitt (88 AD, 88 controls). WebRTC VAD isolates speech segments, and 99 handcrafted temporal–acoustic features including MFCC (with Δ and ΔΔ) are extracted. A speaker-independent evaluation via GroupShuffleSplit reports an average SVM (RBF) AUC of 0.674 ± 0.091.","Transcript-Free Lightweight Detection of Alzheimer's Disease from Spontaneous Speech Using Handcrafted MFCC-Dominant Acoustic  \nBiomarkers  \n1st Rashin Gholijani Farahani  \nDepartment of computer Engineering, Ka.c., Islamic Azad University Karaj,Alborz, Iran  \n[rashin.gholijanifarahani@iau.ir](rashin.gholijanifarahani@iau.ir)  \n2nd Azam Bastanfard  \nCorresponding author Department of computer Engineering, Ka.c., Islamic Azad University Karaj,Alborz ,Iran [bastanfard@iau.ac.ir](bastanfard@iau.ac.ir)  \nAbstract—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 anon-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 astringent speaker-independent GroupShuffleSplit,documenting performance across 30 iterations. A lightweight SVM with an RBF kernel gets an average AUC of 0.674 ± 0.091 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 ± 0.091). 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.  \nKeywords—Alzheimer’s disease; spontaneous speech; acoustic biomarkers; pause analysis; WebRTC VAD; MFCC; transcript-free; speaker-independent evaluation; lightweight machine learning; digital screening  \nI. INTRODUCTION  \nAlzheimer's disease (AD) is a neurodegenerative disorder that gets worse over time and affects memory, executive function, and communication. Since clinical symptoms frequently manifest only after significant neurodegeneration has transpired, there is a persistent necessity for screening instruments that are timely, accessible, and economical. In this context, spontaneous speech has garnered increasing attention as a non-invasive digital biomarker, as it can indicate early disturbances in temporal control, fluency, and speech planning linked to cognitive decline.  \nA significant corpus of evidence demonstrates that temporal speech anomalies characterized by prolonged pauses, elevated silence ratios, hesitation patterns, and diminished speaking duration are correlated with Alzheimer's disease  \n(AD) and mild cognitive impairment (MCI) [1]–[6] . Crosslingual studies in English, Chinese, Swedish, and Spanish indicate that cognitively impaired speakers display pauses that are longer, more frequent, and more variable compared to healthy controls [1], [2], [5], [7] . Additionally, the distributional characteristics of pauses, such as upper-tail statistics (e.g., P90), long-to-short pause ratios, and decreased articulation time, have been identified as useful for differentiating pathological from normative aging [3, 4, 7] .  \nIn addition to timing-related markers, acoustic and spectral cues can offer supplementary diagnostic information. Alterations in vocal energy, rhythm stability, spectral dynamics, and MFCC-based representations have been associated with early cognitive decline [10]–[12] . Recent advancements in deep learning, encompassing convolutional networks, self-supervised speech encoders (e.g., wav2vecstyle models), and multimodal ","cbCaivIl0oRvYKxM","https://ap.wps.com/l/cbCaivIl0oRvYKxM","pdf",525155,1,10,"English","en",105,"# Introduction\n## Problem Motivation\n## Proposed Lightweight Transcript-Free Baseline\n## Feature Extraction and Evaluation","[{\"question\":\"Why is transcript-free detection important for Alzheimer’s disease screening?\",\"answer\":\"Many high-performing approaches rely on transcripts/ASR and language-dependent features, which can introduce errors and reduce usefulness in low-resource settings. A transcript-free audio-only baseline avoids dependence on transcripts and reduces practical constraints.\"},{\"question\":\"What dataset and speech material are used in the method?\",\"answer\":\"The approach uses 176 spontaneous speech recordings from the DementiaBank Pitt corpus, specifically the Cookie Theft task, comprising 88 Alzheimer’s disease cases and 88 controls.\"},{\"question\":\"How is speech segmented and what features are extracted?\",\"answer\":\"WebRTC voice activity detection separates speech from non-speech. The system extracts 99 handcrafted temporal–acoustic features, including pause and fluency statistics, spectral/prosodic descriptors, and MFCC summaries with Δ and ΔΔ.\"}]",1784206555,25,{"code":4,"msg":30,"data":31},"ok",{"site_id":24,"language":23,"slug":32,"title":13,"keywords":33,"description":14,"schema_data":34,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":27},"transcript-free-lightweight-detection-of-alzheimers-disease-from-spontaneous-speech-using-handcrafted-mfcc-dominant-acoustic-biomarkers","",{"@graph":35,"@context":85},[36,53,68],{"@type":37,"itemListElement":38},"BreadcrumbList",[39,43,47,50],{"item":40,"name":41,"@type":42,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":44,"name":45,"@type":42,"position":46},"https://docshare.wps.com/document/","Document",2,{"item":48,"name":12,"@type":42,"position":49},"https://docshare.wps.com/document/research-report/",3,{"item":51,"name":13,"@type":42,"position":52},"https://docshare.wps.com/document/transcript-free-lightweight-detection-of-alzheimers-disease-from-spontaneous-speech-using-handcrafted-mfcc-dominant-acoustic-biomarkers/85831/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":23,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":40,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-17","2026-07-16",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"Why is transcript-free detection important for Alzheimer’s disease screening?","Question",{"text":75,"@type":76},"Many high-performing approaches rely on transcripts/ASR and language-dependent features, which can introduce errors and reduce usefulness in low-resource settings. A transcript-free audio-only baseline avoids dependence on transcripts and reduces practical constraints.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What dataset and speech material are used in the method?",{"text":80,"@type":76},"The approach uses 176 spontaneous speech recordings from the DementiaBank Pitt corpus, specifically the Cookie Theft task, comprising 88 Alzheimer’s disease cases and 88 controls.",{"name":82,"@type":73,"acceptedAnswer":83},"How is speech segmented and what features are extracted?",{"text":84,"@type":76},"WebRTC voice activity detection separates speech from non-speech. The system extracts 99 handcrafted temporal–acoustic features, including pause and fluency statistics, spectral/prosodic descriptors, and MFCC summaries with Δ and ΔΔ.","https://schema.org",{"og:url":51,"og:type":87,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":89,"canonical":51},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,134],{"id":20,"doc_module":4,"doc_module_name":45,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":45,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":45,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":45,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":45,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":45,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":45,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":21,"doc_module":4,"doc_module_name":45,"category_name":132,"show_sort_weight":21,"slug":133},"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":45,"category_name":136,"show_sort_weight":106,"slug":137},19,"General","general"]