Back to top Open menu arrow
Purchase PDF Reprints

The Journal of
Clinical Psychiatry
Members only

The Journal of Clinical Psychiatry

Original Research Focus on Women’s Mental Health September 9, 2026

Using Speech to Develop Multivariable Prediction Models for Major Depressive Disorder and Generalized Anxiety Disorder During Pregnancy

; ; ; ; ; ;

J Clin Psychiatry 2026;87(4):26m16487

Abstract

Objective: Depression and anxiety are common during pregnancy yet remain underdetected. In this project, we examined the usefulness of speech for predicting major depressive disorder (MDD) and generalized anxiety disorder (GAD) during pregnancy.

Methods: We conducted a Prediction Model Development and Evaluation Study using data collected from July 2019 to May 2020 in British Columbia, Canada. MDD and GAD diagnoses were ascertained using the Structured Clinical Interview for DSM-5. We extracted speech features from recorded interviews that contained the voices of both patient and the interviewer (full interview), as well as from manually verified speech segments (VS) of patients. We trained several machine learning models and compared the predictive performance of speech-based models with combined models—consisting of pregnancy/sociodemographic characteristics and speech features—in multivariate analyses.

Results: The total sample included 146 participants: 19 with MDD, 28 with GAD (3 had comorbid MDD and GAD), and 102 controls (no current/past known psychiatric disorders). Speech models based on the full interviews showed consistent patterns of good performance for MDD (F1-scores between 74%–77%) and GAD (F1-scores between 76%–80%). Models based on VS showed poor performance for MDD (F1-scores between 42%–45%) and acceptable performance for GAD (F1-scores between 51%–65%). Adding pregnancy/sociodemographic characteristics to the best speech models did not improve the performance of GAD models (best F1-score 78.00%±19.86) and slightly reduced that of MDD models (best F1-score 76.00%±11.45).

Conclusion: The most influential speech features for prediction of MDD and GAD during pregnancy aligned with those reported among the general population.

J Clin Psychiatry 2026;87(4):26m16487

Author affiliations are listed at the end of this article.

Also from JCP

Electroconvulsive Therapy and All-Cause, Nonsuicide, and Suicide Mortality Obstetric Risk Associated with Stimulant vs Nonstimulant Treatment ADHD Medication Safety in Pregnancy: Precise Estimates, Uncertain Causal Inference

From the Editors

New Series The JCP Podcast Take-home ketamine for depression? Dr. Benjamin Brody calls the answer an unambiguous no, citing self-escalation risk, a narrow therapeutic index, and rare but terrifying dysphoric dissociation. He also unpacks why esketamine's tightly monitored REMS pathway looks nothing like the largely unregulated ketamine market.

Members Only Content

This full article is available exclusively to Professional tier members. Subscribe now to unlock the HTML version and gain unlimited access to our entire library plus all PDFs. If you're already a subscriber, please log in below to continue reading.

Subscribe Now

Already a member? Log in

  1. Melville JL, Gavin A, Guo Y, et al. Depressive disorders during pregnancy: prevalence and risk factors in a large urban sample. Obstet Gynecol. 2010;116(5):1064–1070. PubMed CrossRef
  2. Dennis CL, Falah-Hassani K, Shiri R. Prevalence of antenatal and postnatal anxiety: systematic review and meta-analysis. Br J Psychiatry. 2017;210(5):315–323. PubMed CrossRef
  3. Hagatulah N, Brann E, Oberg AS, et al. Perinatal depression and risk of mortality: nationwide, register based study in Sweden. BMJ. 2024;384:e075462. PubMed CrossRef
  4. Staneva A, Bogossian F, Pritchard M, et al. The effects of maternal depression, anxiety, and perceived stress during pregnancy on preterm birth: a systematic review. Women Birth. 2015;28(3):179–193. PubMed CrossRef
  5. Voit FAC, Kajantie E, Lemola S, et al. Maternal mental health and adverse birth outcomes. PLoS One. 2022;17(8):e0272210. CrossRef
  6. Babineau V, Fonge YN, Miller ES, et al. Associations of maternal prenatal stress and depressive symptoms with childhood neurobehavioral outcomes in the ECHO cohort of the NICHD fetal growth studies: fetal growth velocity as a potential mediator. J Am Acad Child Adolesc Psychiatry. 2022;61(9):1155–1167. PubMed CrossRef
  7. Lin Y, Xu J, Huang J, et al. Effects of prenatal and postnatal maternal emotional stress on toddlers’ cognitive and temperamental development. J Affect Disord. 2017;207:9–17. PubMed CrossRef
  8. Webb R, Uddin N, Constantinou G, et al. Meta-review of the barriers and facilitators to women accessing perinatal mental healthcare. BMJ Open. 2023;13(7):e066703. CrossRef
  9. Bayrampour H, McNeil DA, Benzies K, et al. A qualitative inquiry on pregnant women’s preferences for mental health screening. BMC Pregnancy Childbirth. 2017;17(1):339. PubMed CrossRef
  10. Wakida EK, Talib ZM, Akena D, et al. Barriers and facilitators to the integration of mental health services into primary health care: a systematic review. Syst Rev. 2018;7(1):211. CrossRef
  11. Wang J, Zhang L, Liu T, et al. Acoustic differences between healthy and depressed people: a cross-situation study. BMC Psychiatry. 2019;19(1):300. PubMed CrossRef
  12. Cummins N, Scherer S, Krajewski J, et al. A review of depression and suicide risk assessment using speech analysis. Speech Commun. 2015;71:39. CrossRef
  13. Albuquerque L, Valente ARS, Teixeira A, et al. Association between acoustic speech features and non-severe levels of anxiety and depression symptoms across lifespan. PLoS One. 2021;16(4):e0248842. CrossRef
  14. Low DM, Bentley KH, Ghosh SS. Automated assessment of psychiatric disorders using speech: a systematic review. Laryngoscope Investig Otolaryngol. 2020;5(1):96–116. PubMed CrossRef
  15. Belouali A, Gupta S, Sourirajan V, et al. Acoustic and language analysis of speech for suicidal ideation among US veterans. BioData Min. 2021;14(1):11. CrossRef
  16. Quatieri TF, Malyska N. Vocal-source biomarkers for depression: A link to psychomotor activity; 2012:1059–1062.
  17. Kiss G, Vicsi K. Mono- and multi-lingual depression prediction based on speech processing. Int J Speech Technol. 2017;20(4):919–935. CrossRef
  18. Denes PB, Pinson EN. The Speech Chain: The Physics and Biology of Spoken Language. Bell Telephone Laboratories; 1963.
  19. Enge S, Fleischhauer M, Lesch KP, et al. Serotonergic modulation in executive functioning: linking genetic variations to working memory performance. Neuropsychologia. 2011;49(13):3776–3785. PubMed CrossRef
  20. Dwivedi Y. Brain-derived neurotrophic factor: role in depression and suicide. Neuropsychiatr Dis Treat. 2009;5:433–449. PubMed CrossRef
  21. Brigitta B. Pathophysiology of depression and mechanisms of treatment. Dialogues Clin Neurosci. 2002;4(1):7–20. PubMed
  22. National Institute on Deafness and Other Communication Disorders. Voice, speech, and language: what are they? Retrieved on October 15, 2024 from. https://www.nidcd.nih.gov/sites/default/files/2023-03/voice-speech-language-1.pdf
  23. Behrman A. Speech and Voice Science. 4th ed. vol p.163. Plural Publishing, Inc.; 2023.
  24. Saran M, Georgakopoulos B, Bordoni B. Anatomy, head and neck, larynx vocal cords. StatPearls. 2024.
  25. Koybasi SS, Bicer YO, Kukner A, et al. Effect of pregnancy on vocal cord histology: an animal experiment. Balkan Med J. 2016;33(4):448–452. CrossRef
  26. La FM, Sundberg J. Pregnancy and the singing voice: reports from a case study. J Voice. 2012;26(4):431–439. PubMed CrossRef
  27. Pan W, Deng F, Wang X, et al. Exploring the ability of vocal biomarkers in distinguishing depression from bipolar disorder, schizophrenia, and healthy controls. Front Psychiatry. 2023;14:1079448. CrossRef
  28. Fanos V, Dessì A, Deledda L, et al. Postpartum depression screening through artificial intelligence: preliminary data through the talking about algorithm. J Pediatr Neonatal Individ Med (JPNIM). 2023;12(2):e120222.
  29. Collins GS, Moons KG, Dhiman P, et al. TRIPOD+AI statement: updated guidance for reporting clinical prediction models that use regression or machine learning methods. BMJ. 2024;385:q902. CrossRef
  30. Bayrampour H, Hohn RE, Tamana SK, et al. Pregnancy-Specific Anxiety Tool (PSAT): instrument development and psychometric evaluation. J Clin Psychiatry. 2023;84(3):22m14696. CrossRef
  31. American Psychiatric Association. Structured Clinical Interview for DSM-5 (SCID-5);2015.
  32. Marmar CR, Brown AD, Qian M, et al. Speech-based markers for posttraumatic stress disorder in US veterans. Depress Anxiety. 2019;36(7):607–616. PubMed CrossRef
  33. Pan W, Flint J, Shenhav L, et al. Re-examining the robustness of voice features in predicting depression: compared with baseline of confounders. PLoS One. 2019;14(6):e0218172. PubMed CrossRef
  34. Moore E 2nd, Clements MA, Peifer JW, et al. Critical analysis of the impact of glottal features in the classification of clinical depression in speech. IEEE Trans Biomed Eng. 2008;55(1):96–107. PubMed CrossRef
  35. Bredin H, Yin RQ, Coria JM, et al. Pyannote.Audio: neural building blocks for speaker diarization. Int Conf Acoust Spee. 2020:7124–7128.
  36. Eyben F, Wöllmer M, Schuller B. OpenSmile: the Munich versatile and fast open-source audio feature extractor. In:Proceedings of the 18th ACM International Conference on Multimedia; 2010.
  37. Baevski A, Zhou H, Mohamed A, et al. wav2vec 2.0: a framework for self-supervised learning of speech representations. In: presented at: 34th Conference on Neural Information Processing Systems (NeurIPS 2020). 2020.
  38. Fernandes J, Teixeira F, Guedes V, et al. Harmonic to noise ratio measurement -  selection of window and length. Pr. 2018;138:280–285. CrossRef
  39. Teixeira JP, Oliveira C, Lopes C. Vocal acoustic analysis – Jitter, shimmer and HNR parameters. Procedia Technology. 2013;9:1112–1122. CrossRef
  40. Aloshban N, Esposito A, Vinciarelli A. Detecting depression in less than 10 seconds: impact of speaking time on depression detection sensitivity. In: Proceedings of the 2020 International Conference on Multimodal Interaction. Virtual Event; 2020.
  41. Barandas M, Duarte F, Fernandes L, et al. TSFEL: time series feature extraction library. SoftwareX. 2020;11:100456. CrossRef
  42. scikit-learn developers. StratifiedKFold. https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.StratifiedKFold.html. Accessed Oct 21, 2025
  43. Akiba T, Sano S, Yanase T, et al. Optuna: a next-generation hyperparameter optimization framework. In: Kdd’19: Proceedings of the 25th Acm Sigkdd International Conference on Knowledge Discovery and Data Mining; 2019:2623–2631.
  44. Areosa I, Torgo L. Explaining the performance of black box regression models. Pr Int Conf Data Sc. 2019:110–118.
  45. Areosa I, Torgo L. Visual interpretation of regression error. Expert Syst 2020;37(6). CrossRef
  46. Bauer JF, Gerczuk M, Schindler-Gmelch L, et al. Validation of machine learning-based assessment of major depressive disorder from paralinguistic speech characteristics in routine care. Depress Anxiety. 2024;2024:9667377. CrossRef
  47. Wang JY, Sui XY, Zhu TS, et al. Identifying comorbidities from depressed people via voice analysis. Ieee Int C Bioinform. 2017:986–991.
  48. Lin D, Nazreen T, Rutowski T, et al. Feasibility of a machine learning-based smartphone application in detecting depression and anxiety in a generally senior population. Front Psychol. 2022;13:811517. CrossRef
  49. Rutowski T, Harati A, Lu Y, et al. Optimizing Speech-Input Length for Speaker-Independent Depression Classification; 2019.
  50. Pendse SR, Sharma A, Vashistha A, et al. Can I not be suicidal on a Sunday?. In: Understanding Technology-Mediated Pathways to Mental Health Support. Proc SIGCHI Conf Hum Factor Comput Syst; 2021.
  51. Koutsouleris N, Hauser TU, Skvortsova V, et al. From promise to practice: towards the realisation of AI-informed mental health care. Lancet Digit Health. 2022;4(11):e829–e840. CrossRef
  52. Wiens J, Saria S, Sendak M, et al. Do no harm: a roadmap for responsible machine learning for health care. Nat Med. 2019;25(9):1337–1340. PubMed CrossRef
  53. Arseniev-Koehler A, Mozgai S, Scherer S. What type of happiness are you looking for? - A closer look at detecting mental health from language. Association for Computational Linguistics; 2018:1–12. CrossRef

About the Authors

Buy PDF for $40

Please sign in or purchase this PDF for $40.

The Journal of Clinical Psychiatry June 2026 cover with featured article titles

Vol 87 • 2026 • Number 2

Read the Current Issue

Original Research

CBT following Esketamine for Major Depression and Suicidal Ideation for Relapse Prevention: The CBT-ENDURE Randomized Trial

Original Research

Neuropsychiatric Symptom Clusters and Brain Structure in Alzheimer’s Disease

Systematic Review

Thirty Years of the Diagnostic Criteria for Psychosomatic Research

Original Research

Efficacy of Olanzapine/Samidorphan on Negative Symptoms

Original Research

Comparing Risk Prediction for Suicide Attempts and Deaths After Emergency Department Visits

Clinical and Practical Psychopharmacology

Insomnia, Sleep, and the Use of Sedative-Hypnotic Drugs: Practical Notes
View current Issue

Learn More About Psychedelics in Psychiatry

Explore the Collection
Neural network illustration symbolizing psychedelic research advancements

Subscribe to read
more articles

Subscribe