D.V. Borovikova1, O.V. Grishin2, O.A. Loskutova3, A.V. Yupashevskiy4, A.V. Markov5, A.S. Kazmina6, K.A. Metsler7

1–7 Novosibirsk State Technical University (Novosibirsk, Russia)

In recent years, there has been a sharp increase in the number of the vocal tract functional disorders. As a rule, the diseases causes of this kind are frequent stress and psycho-emotional stress. A timely undetected functional disorder of the vocal tract can lead to serious consequences, and not only reduce social adaptation, but also cause serious organic damage to the vocal folds and other vocal tract organs. Until now, the functional disorders diagnosis is carried out expertly by a phoniatrist or a voice specialist. Such an assessment is subjective and depends on the professional skills of the expert. Alternative diagnostics methods using the acoustic analysis are the subject of criticism by a number of specialists, because it does not show diagnostically significant reliability of research results and directly depend on the speaker individual characteristics.

This paper proposes a new method for diagnosing the voice functional disorders using spectral analysis methods, which does not depend on the speaker physiological characteristics. With a specially developed hardware-software complex, an acoustic analysis of the two most characteristic voice signal samples was carried out in normal conditions and with functional impairment of phonation. The most characteristic differences between the two samples are observed in the values of the normalized amplitudes of the first (more than 7 times), the fourth (more than 2 times), the fifth (more than 4 times), the sixth (more than 13 times) and the seventh (more than 37 times) harmonics. The calculated normalized amplitudes allow to obtain an indicative picture of the voice signal harmonic structure and can be used to identify functional disorders of the voice.

Based on the spectral analysis of the voice signal this work suggests an algorithm that allows to make an objective diagnosis of the voice functional disorders, as well as to assess the effectiveness of therapeutic correction. The developed methodology includes the calculation of the first seven harmonics normalized energies does not take into account the individual characteristics of the speaker. The technique was initially tested on the two most characteristic samples, and significant differences were obtained. To further assess the effectiveness of the presented methodology, it is necessary to conduct research on a larger sample.

Borovikova D.V., Grishin O.V., Loskutova O.A., Yupashevskiy A.V., Markov A.V., Kazmina A.S., Metsler K.A. Development of the method for detecting voice functional disorders by spectral analysis. Biomedicine Radioengineering. 2021. V. 24. № 6. Р. 29−36. DOI: https://doi.org/10.18127/j15604136-202106-03 (In Russian)

1. Vasilenko Yu.S. Golos. Foniatricheskiye aspekty. M.: Dipak. 2013. 396 s. (in Russian).
2. Muzykalnaya entsiklopediya. M.: Sovetskaya entsiklopediya. 1976. T. 3. 1104 s. (in Russian).
3. Fujiki R.B., Thibeault S.L. Examining Relationships Between GRBAS Ratings and Acoustic, Aerodynamic and Patient-Reported Voice Measures in Adults With Voice Disorders // Journal of Voice, 2021. [Elektronnyj dokument]. 2021. — URL: https://www.jvoice.org/ article/S0892-1997(21)00074-6/fulltext
4. Sobol M., Sielska-Badurek E.M. Index (DSI) – Normative Values. Systematic Review and Meta-Analysis // Journal of Voice. 2020. V. 63. P. 951–956.
5. Ataee E., Khoramshahi H., Naderifar E., Dastoor-pour E.M. Relation Between Dysphonia Severity Index (DSI) and Consensus AuditoryPerceptual Evaluation of Voice (CAPE-V) // Journal of Voice. 2020. [Elektronnyj dokument]. — URL:https://www.jvoice.org/ article/S0892-1997(20)30248-4/fulltext
6. Spazzapan E.A., Fabron E.M.G., Berti L., Chagas E.F.B., de Castro Marino V.C. Acoustic Characteristics of the Voice for Brazilian Portuguese Speakers Across the Life Span // Journal of Voice. 2020. DOI: 10.1016/j.jvoice.2020.09.019
7. Fant G. Akusticheskaya teoriya recheobrazovaniya: Per. s angl. / per. L.A. Varshavskiy. V.I. Medvedev. M.: Nauka. 1964. 304 s. (in Russian).
8. Cataldo E., Soize C. Voice Signals Produced With Jitter Through a Stochastic One-mass Mechanical Model // Journal of Voice. 2017.  V. 31. P. 111.e9–111.e18.
9. Barsties B., Latoszek V., Ulozaite-Staniene N., Maryn Y., Petrauskas T., Uloza V. The Influence of Gender and Age on the Acoustic Voice Quality Index and Dysphonia Severity Index: A Normative Study // Journal of Voice. 2019. V. 33(3). P. 340–345.
10. Kim H., Gao S., Yi B., Shi R., Wan Q., Huang Z. Validation of the Dysphonia Severity Index in the Dr. Speech Program // Journal of Voice. 2019. [Elektronnyj dokument]. — URL: https://www.jvoice.org/article/S0892-1997(19)30213-9/fulltext
11. Dwivedi K., Nigam A. Deep Learning Model For Disease Detection By Phoniatrics Biomarkers Using Quantum Computation // 2nd International Conference on Data, Engineering and Applications (IDEA). 2020. [Elektronnyj dokument]. — URL: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=9170711&tag=1