Skip to main content
Springer Nature Link
Log in

Detection of Speech Related Disorders by Pre-trained Embedding Models Extracted Biomarkers

  • Conference paper
  • First Online:
Speech and Computer (SPECOM 2022)

Part of the book series: Lecture Notes in Computer Science ((LNAI,volume 13721))

Included in the following conference series:

Abstract

Several research studies are conducted to support the diagnosis of certain disorders. Depression, Parkinson's disease and dysphonia are such disorders which can manifest in speech. This provides a non-invasive and rapid method to support/confirm the diagnosis. Knowledge-based acoustic features are heavily researched for each disorder. However, the importance and quantity of these features are still open questions. Moreover, this feature-engineering procedure can be time-consuming and may require more effort for analysis. Therefore, it is a state-of-art approach to use the feature extraction part of an out-of-domain speech recognition system for feature extraction. In our research, x-vector and ECAPA pre-trained models were used to derive feature vectors. Binary and multiclass classification were conducted using Support Vector Machines. Nested cross validation method was applied for cost and gamma parameter selection. Our results pointed out that disorders can be recognized with similar accuracy using pre-trained feature extractors as with knowledge-based features in the case of binary classification. This highlights the opportunity to omit feature engineering for every disorder but use the same out-of-domain feature extractor for classification. On the other hand, with four-class classification better results were achieved than in our previous research where knowledge-based features were used. This supports the idea of robust discrimination between disorders.

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

Log in via an institution

Subscribe and save

Springer+ Basic
$34.99 /Month
  • Get 10 units per month
  • Download Article/Chapter or eBook
  • 1 Unit = 1 Article or 1 Chapter
  • Cancel anytime
Subscribe now

Buy Now

Chapter
USD 29.95
Price excludes VAT (USA)
  • Available as PDF
  • Read on any device
  • Instant download
  • Own it forever

Tax calculation will be finalised at checkout

Purchases are for personal use only

Institutional subscriptions

Similar content being viewed by others

References

  1. Robin, J., Harrison, J.E., Kaufman, L.D., Rudzicz, F., Simpson, W., Yancheva, M.: Evaluation of speech-based digital biomarkers: review and recommendations. Digital Biomarkers 4(3), 99–108 (2020). https://doi.org/10.1159/000510820

    Article  Google Scholar 

  2. Ramanarayanan, V., Lammert, A.C., Rowe, H.P., Quatieri, T.F., Green, J.R.: Speech as a biomarker: opportunities, interpretability, and challenges. Perspect. ASHA Spec. Interest Groups 7(1), 276–283 (2022)

    Article  Google Scholar 

  3. Pompili, A., et al.: Automatic detection of parkinson’s disease: an experimental analysis of common speech production tasks used for diagnosis. In: Ekštein, K., Matoušek, V. (eds.) TSD 2017. LNCS (LNAI), vol. 10415, pp. 411–419. Springer, Cham (2017). https://doi.org/10.1007/978-3-319-64206-2_46

    Chapter  Google Scholar 

  4. Liu, Y., Lee, T., Ching, P.C., Law, T.K., Lee, K.Y.: Acoustic assessment of disordered voice with continuous speech based on utterance-level ASR posterior features. IEEE/ACM Trans. Audio, Speech, Lang. Process. 27(6), 1047–1059 (2019)

    Article  Google Scholar 

  5. Vadovsky, M., Paralic, J.: Parkinson’s disease patients classification based on the speech signals. In: 2017 IEEE 15th International Symposium on Applied Machine Intelligence and Informatics (SAMI), pp. 321–326. Herl’any, Slovakia (2017)

    Google Scholar 

  6. Rejaibi, E., Komaty, A., Meriaudeau, F., Agrebi, S., Othmani, A.: MFCC-based recurrent neural network for automatic clinical depression recognition and assessment from speech. Biomed. Signal Process. Control 71, 103107 (2022). https://doi.org/10.1016/j.bspc.2021.103107

    Article  Google Scholar 

  7. Balestrino, R., Schapira, A.H.V.: Parkinson disease. Eur. J. Neurol. 27(1), 27–42 (2020). https://doi.org/10.1111/ene.14108

    Article  Google Scholar 

  8. Mathieson, L.: Green and Mathieson’s the Voice & its Disorders. Whurr Publishers (2001)

    Google Scholar 

  9. Jenei, A.Z., Kiss, G., Tulics, M.G., Sztahó, D.: Separation of several illnesses using correlation structures with convolutional neural networks. Acta Polytech. Hung. 18(7), 47–66 (2021). https://doi.org/10.12700/APH.18.7.2021.7.3

    Article  Google Scholar 

  10. Sztahó, D., et al.: Automatic separation of various disease types by correlation structure of time shifted speech features. In: 2018 41st International Conference on Telecommunications and Signal Processing (TSP). IEEE, pp. 1–4. Greece, Athens (2018)

    Google Scholar 

  11. Sztahó, D., Kiss, G., Tulics, M. G., Vicsi, K.: Automatic discrimination of several types of speech pathologies. In: 2019 International Conference on Speech Technology and Human-Computer Dialogue (SpeD), pp. 1–6. IEEE, Timisoara, Romania (2019)

    Google Scholar 

  12. Sztahó, D., Gábor, K., Miklós, G.T.: Deep learning solution for pathological voice detection using LSTM-based autoencoder hybrid with multi-task learning. In: 14th International Conference on Bio-Inspired Systems and Signal Processing (BIOSIGNALS), pp. 135–141. Vienna, Austria (2021)

    Google Scholar 

  13. Patil, M., Wadhai, V.: Selection of classifiers for depression detection using acoustic features. In: 2021 International Conference on Computational Intelligence and Computing Applications (ICCICA), pp. 1–4. Nagpur, India (2021)

    Google Scholar 

  14. Verde, L., et al.: A lightweight machine learning approach to detect depression from speech analysis. In: 2021 IEEE 33rd International Conference on Tools with Artificial Intelligence (ICTAI), pp. 330–335. Washington, DC, USA (2021)

    Google Scholar 

  15. Braga, D., Madureira, A.M., Coelho, L., Ajith, R.: Automatic detection of Parkinson’s disease based on acoustic analysis of speech. Eng. Appl. Artif. Intell. 7, 148–158 (2019)

    Article  Google Scholar 

  16. Umapathy, S., Rachel, S., Thulasi, R.: Automated speech signal analysis based on feature extraction and classification of spasmodic dysphonia: a performance comparison of different classifiers. Int. J. Speech Technol. 21(1), 9–18 (2017). https://doi.org/10.1007/s10772-017-9471-8

    Article  Google Scholar 

  17. Harati, A., et al.: Speech-based depression prediction using encoder-weight-only transfer learning and a large corpus. In: 2021 IEEE International Conference on Acoustics. Speech and Signal Processing (ICASSP), pp. 7273–7277. ON, Canada, Toronto (2021)

    Google Scholar 

  18. Botelho, C., Teixeira, F., Rolland, T., Abad, A., Trancoso, I.: Pathological speech detection using x-vector embeddings. arXiv preprint arXiv:2003.00864 (2020)

  19. Egas-López, J.V., Kiss, G., Sztahó, D., Gosztolya, G.: Automatic assessment of the degree of clinical depression from speech using X-vectors. In: 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 8502–8506. Singapore (2022)

    Google Scholar 

  20. Jeancolas, L., et al.: X-Vectors: new quantitative biomarkers for early Parkinson’s disease detection from speech. Front. Neuroinform. 15, 578369 (2021)

    Article  Google Scholar 

  21. Snyder, D., Garcia-Romero, D., Povey, D., Khudanpur, S.: Deep neural network embeddings for text-independent speaker verification. In: Interspeech 2017, pp. 999–1003. Stockholm, Sweden (2017)

    Google Scholar 

  22. Desplanques, B., Thienpondt, J., Demuynck, K.: Ecapa-tdnn: Emphasized channel attention, propagation and aggregation in tdnn based speaker verification. arXiv preprint arXiv:2005.07143. (2020)

  23. Cortes, C., Vapnik, V.: Support vector networks. Mach. Learn. 20(3), 273–297 (1995)

    Article  MATH  Google Scholar 

  24. Chang, C.-C., Lin, C.-J.: LIBSVM: a library for support vector machines. ACM Trans. Intel. Syst. Technol. (TIST) 2(3), 1–27 (2011)

    Article  Google Scholar 

  25. Gosztolya, G., Vincze, V., Tóth, L., Pákáski, M., Kálmán, J., Hoffmann, I.: Identifying mild cognitive impairment and mild Alzheimer’s disease based on spontaneous speech using ASR and linguistic features. Comput. Speech Lang. 53, 181–197 (2019)

    Article  Google Scholar 

  26. Beck, A.T., Steer, R.A., Ball, R., Ranieri, W.F.: Comparison of beck depression inventories -IA and -II in psychiatric outpatients. J. Pers. Assess. 67(3), 588–597 (1996)

    Article  Google Scholar 

  27. Hoehn, M., Yahr, M.D.: Parkinsonism onset, progression, and mortality. Neurology 17(5), 427–442 (1967)

    Article  Google Scholar 

  28. Gaber, A.G.H., Liang, F.-Y., Yang, J.-S., Wang, Y.-J., Zheng, Y.-Q.: Correlation among the dysphonia severity index (DSI), the RBH voice perceptual evaluation, and minimum glottal area in female patients with vocal fold nodules. J. Voice 28(1), 20–23 (2014)

    Article  Google Scholar 

  29. Chung, J. S., Nagrani, A., Zisserman, A.: VoxCeleb2: deep speaker recognition. In: Proceedings of the Interspeech 2018, pp. 1086–1090. Hyderabad, India (2018)

    Google Scholar 

Download references

Acknowledgements

The work was funded by project no. K128568 that has been implemented with the support provided from the National Research, Development and Innovation Fund of Hungary, financed under the K_18 funding scheme.

Author information

Authors and Affiliations

  1. Budapest University of Technology and Economics, Magyar tudósok körútja 2, Budapest, 1117, Hungary

    Attila Zoltán Jenei, Gábor Kiss & Dávid Sztahó

Authors
  1. Attila Zoltán Jenei
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Gábor Kiss
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Dávid Sztahó
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Dávid Sztahó .

Editor information

Editors and Affiliations

  1. Indian Institute of Technology Dharwad, Dharwad, India

    S. R. Mahadeva Prasanna

  2. St. Petersburg Federal Research Center of the Russian Academy of Sciences, St. Petersburg, Russia

    Alexey Karpov

  3. Koneru Lakshmaiah Education Foundation, Vaddeswaram, India

    K. Samudravijaya

  4. KIIT Group of Colleges, Gurugram, India

    Shyam S. Agrawal

Rights and permissions

Reprints and permissions

Copyright information

© 2022 Springer Nature Switzerland AG

About this paper

Check for updates. Verify currency and authenticity via CrossMark

Cite this paper

Jenei, A.Z., Kiss, G., Sztahó, D. (2022). Detection of Speech Related Disorders by Pre-trained Embedding Models Extracted Biomarkers. In: Prasanna, S.R.M., Karpov, A., Samudravijaya, K., Agrawal, S.S. (eds) Speech and Computer. SPECOM 2022. Lecture Notes in Computer Science(), vol 13721. Springer, Cham. https://doi.org/10.1007/978-3-031-20980-2_24

Download citation

  • DOI: https://doi.org/10.1007/978-3-031-20980-2_24

  • Published:

  • Publisher Name: Springer, Cham

  • Print ISBN: 978-3-031-20979-6

  • Online ISBN: 978-3-031-20980-2

  • eBook Packages: Computer ScienceComputer Science (R0)

Publish with us

Policies and ethics