|

Identification of the Pilot as Part of the Crew Using Speech Spectral Transfer Function

Authors: Korsun O.N., Mikhaylov E.I. Published: 12.10.2019
Published in issue: #5(128)/2019  
DOI: 10.18698/0236-3933-2019-5-35-48

 
Category: Informatics, Computer Engineering and Control | Chapter: System Analysis, Control, and Information Processing  
Keywords: human-operator identification, speech spectral transfer function, frequency, classification, k nearest neighbors method

The paper deals with the problem of voice identification of the pilot as part of the crew, which is one of the ways to improve the interface of the cockpit of a modern aircraft. The main trends of pilot voice identification in the task of improving the cockpit interface are voice control of onboard equipment and accident investigation. We introduce a method for identifying the speakers personality using the speakers voice transfer function by frequency and the k-nearest neighbors data classification algorithm. Due to the nature of the task, identification was carried out for small groups of operators of up to four people. The main results of testing the proposed method on the experimental speech data that include 3 and 20 different isolated words are given. Findings of research show that the operator can be identified by a small number of code words with an accuracy of about 97--99 % when using the speakers voice transfer function by frequency. The paper also presents a comparison of the results of applying the methodology for pilots of helicopter aviation with the diagnosis of hearing loss and for a group of operators without diseases of the organs of hearing

This work was supported by the Russian Foundation for Basic Research (RFBR project no. 18-08-01142-a)

References

[1] Zheltov S.Yu., Sebryakov G.G. [Simulation of multi-functional human-operator activity while driving aircraft]. XXVII Vseros. nauch.-tekh. konf. shkoly-seminara “Peredacha, priem, obrabotka i otobrazhenie informatsii o bystroprotekayushchikh protsessakh” [XXVII Russ. Sci.-Tech. Conf. of School-Seminar “Transmission, receiving, processing and displaying information on high-speed processes”]. Moscow, ID Akademii im. N.E. Zhukovskogo Publ., 2016, pp. 266--275 (in Russ.).

[2] Rabiner L., Juang B.H. Fundamentals of speech recognition. Prentice Hall, 1993.

[3] Agranovskiy A.B., Lednov D.A., Repalov S.A. Method of text-independent speaker identification based on individuality of vowels pronouncing. Akustika i prikladnaya lingvistika, 2002, no. 3, pp. 103--115 (in Russ.).

[4] Agranovskiy A.V., Lednov D.A. Teoreticheskie aspekty algoritmov obrabotki i klassifikatsii rechevykh signalov [Theoretical aspects of processing and classification algorithms for speech signals]. Moscow, Radio i svyaz Publ., 2004.

[5] Zubova P.I., Koval S.L. Person identification by voice and speech based on complex analysis of phonograms. Teoriya i praktika sudebnoy ekspertizy [Theory and Practice of Forensic Science], 2007, no. 3 (7), pp. 68--76 (in Russ.).

[6] Sorokin V.N., Vyugin V.V., Tananykin A.A. Person recognition by voice: analytical review. Informatsionnye protsessy [Information Processes], 2012, vol. 12, no. 1 (in Russ.). Available at: http://www.jip.ru/2012/1-30-2012.pdf

[7] Greenberg C.S., Martin A.F., Przybocki M.A. Human assisted speaker recognition. Available at: https://www.nist.gov/sites/default/files/documents/oles/8-Greenberg_Craig-Human-Assisted-Speaker-Recognition.pdf (accessed: 15.02.2019).

[8] Evstigneev D.A., Kopysov V.Kh. Psikhologicheskie osnovy vzaimodeystviya v ekipazhe [Psychological basis of interaction in the crew]. Ulyanovsk, UVAU GA Publ., 2007.

[9] Korsun O.N., Mikhaylov E.I., Nakhaev M.Z. Speech spectral transfer function. ITM Web Conf., 2017, vol. 10. DOI: https://doi.org/10.1051/itmconf/20171001006

[10] Korsun O.N., Nakhaev M.Z. The dictor speech transfer function in frequency domain. Vestnik komp’yuternykh i informatsionnykh tekhnologiy [Herald of Computer and Information Technologies], 2016, no. 10 (148), pp. 11--16 (in Russ.).

[11] Korsun O.N., Mikhaylov E.I. The method for speech detection based on control charts techniques. Vestnik komp’yuternykh i informatsionnykh tekhnologiy [Herald of Computer and Information Technologies], 2017, no. 9 (159), pp. 24--31 (in Russ.).

[12] Korsun O.N., Nakhaev M.Z. Concept of speaker speech transfer function and experimental assessment of its properties. Trudy GosNIIAS. Ser. Voprosy avioniki, 2016, no. 1 (25), pp. 33--38 (in Russ.).

[13] Hastie T., Tibshirani R., Friedman J. The elements of statistical learning. Springer Series in Statistics. New York, Springer, 2009. DOI: 10.1007/978-0-387-84858-7

[14] Nigsch F., Bender A., van Buuren B., et al. Melting point prediction employing k-nearest neighbor algorithms and genetic parameter optimization. J. Chem. Inf. Model., 2006, vol. 46, no. 6, pp. 2412--2422. DOI: https://doi.org/10.1021/ci060149f

[15] Vorontsov K.V. Algoritmy klasterizatsii i mnogomernogo shkalirovaniya [Clusterization and multidimensional scaling algorithms]. Moscow, MSU Publ., 2007.

[16] Korsun O.N., Mikhaylov E.I., Nakhaev M.Z. Rezultaty issledovaniy parametrov rechi letchikov vertoletnoy aviatsii [Results of studies on speech parameters of helicopter pilots]. Nauchnye chteniya po aviatsii, posvyashchennye pamyati N.E. Zhukovskogo [Scientific readings on aviation dedicated to memory of N.E. Zhukovsky], 2014, no. 2, pp. 158--161 (in Russ.).

[17] Korsun O.N., Ivanov A.I., Mikhaylov E.I. [Study on pilots speech features with "hardness of hearing" diacrisis]. XXIV Vseros. nauch.-tekh. konf. shkoly-seminara "Peredacha, priem, obrabotka i otobrazhenie informatsii o bystroprotekayushchikh protsessakh" [XXIV Russ. Sci.-Tech. Conf. of School-Seminar "Transmission, receiving, processing and displaying information on high-speed processes"]. Moscow, RPA APR Publ., 2013, pp. 428--438 (in Russ.).