{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,12]],"date-time":"2026-10-12T21:30:56Z","timestamp":1791840656736,"version":"4.3.4"},"reference-count":42,"publisher":"Public Library of Science (PLoS)","issue":"12","license":[{"start":{"date-parts":[[2022,12,22]],"date-time":"2022-12-22T00:00:00Z","timestamp":1671667200000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["www.plosdigitalhealth.org"],"crossmark-restriction":false},"short-container-title":["PLOS Digit Health"],"abstract":"<jats:p>Language impairment is an important biomarker of neurodegenerative disorders such as Alzheimer\u2019s disease (AD). Artificial intelligence (AI), particularly natural language processing (NLP), has recently been increasingly used for early prediction of AD through speech. Yet, relatively few studies exist on using large language models, especially GPT-3, to aid in the early diagnosis of dementia. In this work, we show for the first time that GPT-3 can be utilized to predict dementia from spontaneous speech. Specifically, we leverage the vast semantic knowledge encoded in the GPT-3 model to generate text embedding, a vector representation of the transcribed text from speech, that captures the semantic meaning of the input. We demonstrate that the text embedding can be reliably used to (1) distinguish individuals with AD from healthy controls, and (2) infer the subject\u2019s cognitive testing score, both solely based on speech data. We further show that text embedding considerably outperforms the conventional acoustic feature-based approach and even performs competitively with prevailing fine-tuned models. Together, our results suggest that GPT-3 based text embedding is a viable approach for AD assessment directly from speech and has the potential to improve early diagnosis of dementia.<\/jats:p>","DOI":"10.1371\/journal.pdig.0000168","type":"journal-article","created":{"date-parts":[[2022,12,22]],"date-time":"2022-12-22T13:29:38Z","timestamp":1671715778000},"page":"e0000168","update-policy":"https:\/\/doi.org\/10.1371\/journal.pdig.corrections_policy","source":"Crossref","is-referenced-by-count":144,"title":["Predicting dementia from spontaneous speech using large language models"],"prefix":"10.1371","volume":"1","author":[{"given":"Felix","family":"Agbavor","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3805-1837","authenticated-orcid":true,"given":"Hualou","family":"Liang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2022,12,22]]},"reference":[{"issue":"3","key":"pdig.0000168.ref001","doi-asserted-by":"crossref","first-page":"327","DOI":"10.1002\/alz.12328","article-title":"2021 Alzheimer\u2019s disease facts and figures","volume":"17","year":"2021","journal-title":"Alzheimers Dement"},{"key":"pdig.0000168.ref002","doi-asserted-by":"crossref","first-page":"1179573520907397","DOI":"10.1177\/1179573520907397","article-title":"Current and future treatments in Alzheimer disease: an update.","volume":"12","author":"KG Yiannopoulou","year":"2020","journal-title":"J Cent Nerv Syst Dis"},{"issue":"3","key":"pdig.0000168.ref003","doi-asserted-by":"crossref","first-page":"189","DOI":"10.1016\/0022-3956(75)90026-6","article-title":"\u201cMini-mental state\u201d: a practical method for grading the cognitive state of patients for the clinician.","volume":"12","author":"MF Folstein","year":"1975","journal-title":"J Psychiatr Res"},{"issue":"10","key":"pdig.0000168.ref004","doi-asserted-by":"crossref","first-page":"e0222446","DOI":"10.1371\/journal.pone.0222446","article-title":"Early diagnosis of Alzheimer\u2019s disease using combined features from voxel-based morphometry and cortical, subcortical, and hippocampus regions of MRI T1 brain images.","volume":"14","author":"Y Gupta","year":"2019","journal-title":"PLOS ONE"},{"issue":"2","key":"pdig.0000168.ref005","article-title":"Mini-Cog for the diagnosis of Alzheimer\u2019s disease dementia and other dementias within a primary care setting.","author":"DP Seitz","year":"2018","journal-title":"Cochrane Database Syst Rev [Internet]."},{"key":"pdig.0000168.ref006","volume-title":"BDAE: The Boston Diagnostic Aphasia Examination.","author":"H Goodglass","year":"2001"},{"issue":"2","key":"pdig.0000168.ref007","doi-asserted-by":"crossref","first-page":"282","DOI":"10.1109\/JSTSP.2019.2952087","article-title":"A review of automated speech and language features for assessment of cognitive and thought disorders","volume":"14","author":"R Voleti","year":"2019","journal-title":"IEEE J Sel Top Signal Process"},{"key":"pdig.0000168.ref008","article-title":"Comparing Acoustic-based Approaches for Alzheimer\u2019s Disease Detection [Internet].","author":"A Balagopalan","year":"2021","journal-title":"arXiv"},{"issue":"4","key":"pdig.0000168.ref009","doi-asserted-by":"crossref","first-page":"1547","DOI":"10.3233\/JAD-200888","article-title":"Artificial Intelligence, Speech, and Language Processing Approaches to Monitoring Alzheimer\u2019s Disease: A Systematic Review.","volume":"78","author":"S de la Fuente Garcia","year":"2020","journal-title":"J Alzheimers Dis"},{"key":"pdig.0000168.ref010","doi-asserted-by":"crossref","first-page":"100583","DOI":"10.1016\/j.eclinm.2020.100583","article-title":"Linguistic markers predict onset of Alzheimer\u2019s disease.","volume":"28","author":"E Eyigoz","year":"2020","journal-title":"EClinicalMedicine."},{"issue":"2","key":"pdig.0000168.ref011","doi-asserted-by":"crossref","first-page":"407","DOI":"10.3233\/JAD-150520","article-title":"Linguistic Features Identify Alzheimer\u2019s Disease in Narrative Speech","volume":"49","author":"KC Fraser","year":"2016","journal-title":"J Alzheimers Dis"},{"issue":"2","key":"pdig.0000168.ref012","doi-asserted-by":"crossref","first-page":"272","DOI":"10.1109\/JSTSP.2019.2955022","article-title":"An Assessment of Paralinguistic Acoustic Features for Detection of Alzheimer\u2019s Dementia in Spontaneous Speech","volume":"14","author":"F Haider","year":"2020","journal-title":"IEEE J Sel Top Signal Process"},{"key":"pdig.0000168.ref013","doi-asserted-by":"crossref","first-page":"101181","DOI":"10.1016\/j.csl.2020.101181","article-title":"Exploring neural models for predicting dementia from language","volume":"68","author":"W Kong","year":"2021","journal-title":"Comput Speech Lang"},{"issue":"1","key":"pdig.0000168.ref014","first-page":"112","article-title":"Automatic speech analysis for the assessment of patients with predementia and Alzheimer\u2019s disease.","volume":"1","author":"A K\u00f6nig","year":"2015","journal-title":"Alzheimers Dement Diagn Assess Dis Monit."},{"key":"pdig.0000168.ref015","article-title":"Alzheimer\u2019s Dementia Recognition through Spontaneous Speech: The ADReSS Challenge [Internet].","author":"S Luz","year":"2020","journal-title":"arXiv"},{"key":"pdig.0000168.ref016","article-title":"Learning Linguistic Biomarkers for Predicting Mild Cognitive Impairment using Compound Skip-grams [Internet].","author":"SO Orimaye","year":"2015","journal-title":"arXiv"},{"issue":"1","key":"pdig.0000168.ref017","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1186\/s12859-016-1456-0","article-title":"Predicting probable Alzheimer\u2019s disease using linguistic deficits and biomarkers","volume":"18","author":"SO Orimaye","year":"2017","journal-title":"BMC Bioinformatics"},{"key":"pdig.0000168.ref018","volume-title":"SoundNet: Learning Sound Representations from Unlabeled Video. In: Advances in Neural Information Processing Systems [Internet].","author":"Y Aytar","year":"2016"},{"key":"pdig.0000168.ref019","doi-asserted-by":"crossref","unstructured":"Hershey S, Chaudhuri S, Ellis DPW, Gemmeke JF, Jansen A, Moore RC, et al. CNN architectures for large-scale audio classification. In: 2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). 2017. p. 131\u20135.","DOI":"10.1109\/ICASSP.2017.7952132"},{"key":"pdig.0000168.ref020","article-title":"To BERT or Not To BERT: Comparing Speech and Language-based Approaches for Alzheimer\u2019s Disease Detection [Internet].","author":"A Balagopalan","year":"2020","journal-title":"arXiv"},{"key":"pdig.0000168.ref021","article-title":"Detecting cognitive decline using speech only: The ADReSSo Challenge.","author":"S Luz","year":"2021","journal-title":"ArXiv Prepr ArXiv210409356."},{"key":"pdig.0000168.ref022","article-title":"Automated detection of mild cognitive impairment and dementia from voice recordings: A natural language processing approach.","author":"S Amini","year":"2022","journal-title":"Alzheimers Dement [Internet]."},{"key":"pdig.0000168.ref023","article-title":"BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding [Internet].","author":"J Devlin","year":"2019","journal-title":"arXiv"},{"key":"pdig.0000168.ref024","author":"A Radford","year":"2018","journal-title":"Improving language understanding by generative pre-training"},{"issue":"140","key":"pdig.0000168.ref025","first-page":"1","article-title":"Exploring the limits of transfer learning with a unified text-to-text transformer.","volume":"21","author":"C Raffel","year":"2020","journal-title":"J Mach Learn Res"},{"key":"pdig.0000168.ref026","first-page":"1877","volume-title":"Language Models are Few-Shot Learners. In: Advances in Neural Information Processing Systems [Internet].","author":"T Brown","year":"2020"},{"key":"pdig.0000168.ref027","article-title":"Text and Code Embeddings by Contrastive Pre-Training [Internet].","author":"A Neelakantan","year":"2022","journal-title":"arXiv"},{"key":"pdig.0000168.ref028","article-title":"Efficient Estimation of Word Representations in Vector Space [Internet].","author":"T Mikolov","year":"2013","journal-title":"arXiv"},{"key":"pdig.0000168.ref029","first-page":"1532","article-title":"Glove: Global vectors for word representation.","author":"J Pennington","year":"2014","journal-title":"In: Proceedings of the 2014 conference on empirical methods in natural language processing (EMNLP)."},{"key":"pdig.0000168.ref030","article-title":"Deep contextualized word representations [Internet].","author":"ME Peters","year":"2018","journal-title":"arXiv"},{"key":"pdig.0000168.ref031","first-page":"3810","article-title":"Using the Outputs of Different Automatic Speech Recognition Paradigms for Acoustic- and BERT-Based Alzheimer\u2019s Dementia Detection Through Spontaneous Speech.","author":"Y Pan","year":"2021","journal-title":"In: Interspeech 2021 [Internet]."},{"key":"pdig.0000168.ref032","article-title":"Bert: Pre-training of deep bidirectional transformers for language understanding.","author":"J Devlin","year":"2018","journal-title":"ArXiv Prepr ArXiv181004805."},{"issue":"1","key":"pdig.0000168.ref033","doi-asserted-by":"crossref","first-page":"315","DOI":"10.3233\/JAD-210684","article-title":"Combining multimodal behavioral data of gait, speech, and drawing for classification of Alzheimer\u2019s disease and mild cognitive impairment","volume":"84","author":"Y Yamada","year":"2021","journal-title":"J Alzheimers Dis"},{"issue":"4","key":"pdig.0000168.ref034","doi-asserted-by":"crossref","first-page":"041503","DOI":"10.1063\/5.0011697","article-title":"Artificial intelligence for brain diseases: a systematic review","volume":"4","author":"A Segato","year":"2020","journal-title":"APL Bioeng"},{"issue":"6","key":"pdig.0000168.ref035","doi-asserted-by":"crossref","first-page":"585","DOI":"10.1001\/archneur.1994.00540180063015","article-title":"The Natural History of Alzheimer\u2019s Disease: Description of Study Cohort and Accuracy of Diagnosis","volume":"51","author":"JT Becker","year":"1994","journal-title":"Arch Neurol"},{"key":"pdig.0000168.ref036","first-page":"12449","article-title":"wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations","author":"A Baevski","year":"2020","journal-title":"In: Advances in Neural Information Processing Systems [Internet]."},{"key":"pdig.0000168.ref037","doi-asserted-by":"crossref","unstructured":"Wolf T, Debut L, Sanh V, Chaumond J, Delangue C, Moi A, et al. Transformers: State-of-the-Art Natural Language Processing. In: Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing: System Demonstrations [Internet]. Online: Association for Computational Linguistics; 2020 [cited 2022 Jul 14]. p. 38\u201345. Available from: https:\/\/aclanthology.org\/2020.emnlp-demos.6","DOI":"10.18653\/v1\/2020.emnlp-demos.6"},{"key":"pdig.0000168.ref038","first-page":"18","article-title":"librosa: Audio and music signal analysis in python.","author":"B McFee","year":"2015","journal-title":"In: Proceedings of the 14th python in science conference."},{"key":"pdig.0000168.ref039","first-page":"1459","volume-title":"Opensmile: the munich versatile and fast open-source audio feature extractor. In: Proceedings of the 18th ACM international conference on Multimedia [Internet].","author":"F Eyben","year":"2010"},{"issue":"2","key":"pdig.0000168.ref040","doi-asserted-by":"crossref","first-page":"190","DOI":"10.1109\/TAFFC.2015.2457417","article-title":"The Geneva minimalistic acoustic parameter set (GeMAPS) for voice research and affective computing.","volume":"7","author":"F Eyben","year":"2015","journal-title":"IEEE Trans Affect Comput"},{"key":"pdig.0000168.ref041","doi-asserted-by":"crossref","first-page":"624488","DOI":"10.3389\/fcomp.2020.624488","article-title":"Pauses for detection of Alzheimer\u2019s disease.","volume":"2","author":"J Yuan","year":"2021","journal-title":"Front Comput Sci"},{"key":"pdig.0000168.ref042","first-page":"2825","article-title":"Scikit-learn: Machine learning in Python.","volume":"12","author":"F Pedregosa","year":"2011","journal-title":"J Mach Learn Res"}],"container-title":["PLOS Digital Health"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pdig.0000168","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,12,22]],"date-time":"2022-12-22T13:30:06Z","timestamp":1671715806000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pdig.0000168"}},"subtitle":[],"editor":[{"given":"Benjamin P.","family":"Geisler","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2022,12,22]]},"references-count":42,"journal-issue":{"issue":"12","published-online":{"date-parts":[[2022,12,22]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pdig.0000168","relation":{},"ISSN":["2767-3170"],"issn-type":[{"value":"2767-3170","type":"electronic"}],"subject":[],"published":{"date-parts":[[2022,12,22]]}}}