Acoustic Representation Learning for Early Detection of Parkinson's Disease Through Vocal Expression and Speech Pattern Analysis

Authors

  • Cloe Luque Department of Computer Science, University of New Hampshire, Durham, NH, USA.
  • Te Koha Kenny Department of Computer Science, University of Alabama at Birmingham, Birmingham, AL, USA.
  • Millie Stratford School of Computing, Clemson University, Clemson, SC, USA.

Keywords:

Parkinson's disease, acoustic representation learning, speech biomarkers, self-supervised learning, domain generalization, clinical decision support

Abstract

Early detection of Parkinson's disease remains difficult because motor signs emerge after substantial dopaminergic loss and because speech changes are subtle, variable, and easily confounded by speaker identity, recording conditions, language, and task. Acoustic analysis has long suggested that hypophonia, monotonicity, imprecise articulation, and altered timing carry diagnostic information, yet handcrafted features often generalize poorly across cohorts and devices. This manuscript develops a methodological framework for acoustic representation learning that treats vocal expression and speech patterning as structured, multi-task signals rather than as isolated scalar biomarkers. The central contribution is a clinically constrained learning design that combines self-supervised spectrogram pretraining, supervised Parkinson's disease prediction, auxiliary speech tasks, domain-invariance penalties, calibrated uncertainty, and an abstention-aware decision rule. The framework specifies how waveform inputs are transformed into acoustic representations, how pathology-relevant variation can be separated from nuisance variation, and how representation quality should be evaluated with speaker-disjoint and site-held-out validation. Analytical attention is given to failure mechanisms, including shortcut learning from microphones or silence, label ambiguity in prodromal disease, medication-state effects, and the risk that excessive domain invariance removes clinically meaningful signal. The paper also defines an evaluation logic based on external validation, calibration, decision-curve analysis, and transparent reporting. The argument is that early vocal screening requires representations that are sensitive to early dysarthric change, stable across acquisition conditions, and linked to interpretable acoustic evidence. Such representations can support clinician-in-the-loop risk stratification while avoiding claims of autonomous diagnosis.

References

1. Jankovic, J. (2008). Parkinson's disease: Clinical features and diagnosis. Journal of Neurology, Neurosurgery & Psychiatry, 79(4), 368–376. https://doi.org/10.1136/jnnp.2007.131045

2. Postuma, R. B., Berg, D., Stern, M., Poewe, W., Olanow, C. W., Oertel, W., Obeso, J., Marek, K., Litvan, I., Lang, A. E., Halliday, G., Goetz, C. G., Gasser, T., Dubois, B., Chan, P., Bloem, B. R., Adler, C. H., & Deuschl, G. (2015). MDS clinical diagnostic criteria for Parkinson's disease. Movement Disorders, 30(12), 1591–1601. https://doi.org/10.1002/mds.26424

3. Rusz, J., Cmejla, R., Ruzickova, H., & Ruzicka, E. (2011). Quantitative acoustic measurements for characterization of speech and voice disorders in early untreated Parkinson's disease. The Journal of the Acoustical Society of America, 129(1), 350–367. https://doi.org/10.1121/1.3514381

4. Sapir, S., Ramig, L. O., & Fox, C. M. (2008). Speech and swallowing disorders in Parkinson disease. Current Opinion in Otolaryngology & Head and Neck Surgery, 16(3), 205–210. https://doi.org/10.1097/MOO.0b013e3282febd3a

5. Little, M. A., McSharry, P. E., Hunter, E. J., Spielman, J., & Ramig, L. O. (2009). Suitability of dysphonia measurements for telemonitoring of Parkinson's disease. IEEE Transactions on Biomedical Engineering, 56(4), 1015–1022. https://doi.org/10.1109/TBME.2008.2005954

6. Tsanas, A., Little, M. A., McSharry, P. E., & Ramig, L. O. (2010). Accurate telemonitoring of Parkinson's disease progression by noninvasive speech tests. IEEE Transactions on Biomedical Engineering, 57(4), 884–893. https://doi.org/10.1109/TBME.2009.2036000

7. Orozco-Arroyave, J. R., Hönig, F., Arias-Londoño, J. D., Vargas-Bonilla, J. F., Daqrouq, K., Skodda, S., Rusz, J., & Nöth, E. (2016). Automatic detection of Parkinson's disease in running speech spoken in three different languages. The Journal of the Acoustical Society of America, 139(1), 481–500. https://doi.org/10.1121/1.4939739

8. Zhang, Y., Huang, R., Li, R., He, J., Xia, Y., Chen, F., ... & Zhao, Z. (2024, March). Stylesinger: Style transfer for out-of-domain singing voice synthesis. In Proceedings of the AAAI Conference on Artificial Intelligence (Vol. 38, No. 17, pp. 19597-19605).

9. Baevski, A., Zhou, Y., Mohamed, A., & Auli, M. (2020). wav2vec 2.0: A framework for self-supervised learning of speech representations. In Advances in Neural Information Processing Systems (Vol. 33, pp. 12449–12460).

10. Hsu, W.-N., Bolte, B., Tsai, Y.-H. H., Lakhotia, K., Salakhutdinov, R., & Mohamed, A. (2021). HuBERT: Self-supervised speech representation learning by masked prediction of hidden units. IEEE/ACM Transactions on Audio, Speech, and Language Processing, 29, 3451–3460. https://doi.org/10.1109/TASLP.2021.3122291

11. van den Oord, A., Li, Y., & Vinyals, O. (2018). Representation learning with contrastive predictive coding. arXiv preprint arXiv:1807.03748.

12. Ganin, Y., Ustinova, E., Ajakan, H., Germain, P., Larochelle, H., Laviolette, F., Marchand, M., & Lempitsky, V. (2016). Domain-adversarial training of neural networks. Journal of Machine Learning Research, 17(59), 1–35.

13. Arjovsky, M., Bottou, L., Gulrajani, I., & Lopez-Paz, D. (2019). Invariant risk minimization. arXiv preprint arXiv:1907.02893.

14. Guo, C., Pleiss, G., Sun, Y., & Weinberger, K. Q. (2017). On calibration of modern neural networks. In Proceedings of the 34th International Conference on Machine Learning (Vol. 70, pp. 1321–1330).

15. Kendall, A., & Gal, Y. (2017). What uncertainties do we need in Bayesian deep learning for computer vision? In Advances in Neural Information Processing Systems (Vol. 30, pp. 5574–5584).

16. Vickers, A. J., & Elkin, E. B. (2006). Decision curve analysis: A novel method for evaluating prediction models. Medical Decision Making, 26(6), 565–574. https://doi.org/10.1177/0272989X06295361

17. Collins, G. S., Reitsma, J. B., Altman, D. G., & Moons, K. G. M. (2015). Transparent reporting of a multivariable prediction model for individual prognosis or diagnosis (TRIPOD): The TRIPOD statement. Annals of Internal Medicine, 162(1), 55–63. https://doi.org/10.7326/M14-0697

18. Rusz, J., Cmejla, R., Tykalova, T., Ruzickova, H., Klempir, J., Majerova, V., & Ruzicka, E. (2013). Imprecise vowel articulation as a potential early marker of Parkinson's disease: Effect of speaking task. The Journal of the Acoustical Society of America, 134(3), 2171–2181. https://doi.org/10.1121/1.4816541

19. Goetz, C. G., Tilley, B. C., Shaftman, S. R., Stebbins, G. T., Fahn, S., Martinez-Martin, P., Poewe, W., Sampaio, C., Stern, M. B., Dodel, R., Dubois, B., Holloway, R., Jankovic, J., Kulisevsky, J., Lang, A. E., Lees, A., Lehericy, S., Lewitt, P. A., Nyenhuis, D., ... & LaPelle, N. (2008). Movement Disorder Society-sponsored revision of the Unified Parkinson's Disease Rating Scale (MDS-UPDRS): Scale presentation and clinimetric testing results. Movement Disorders, 23(15), 2129–2170. https://doi.org/10.1002/mds.22340

20. Topol, E. J. (2019). High-performance medicine: The convergence of human and artificial intelligence. Nature Medicine, 25(1), 44–56. https://doi.org/10.1038/s41591-018-0300-7

Downloads

Published

2026-09-23

How to Cite

Cloe Luque, Te Koha Kenny, & Millie Stratford. (2026). Acoustic Representation Learning for Early Detection of Parkinson’s Disease Through Vocal Expression and Speech Pattern Analysis. International Journal of Clinical and Translational Medicine, 2(2). Retrieved from https://ijctmed.org/index.php/home/article/view/237