Machine learning models for student aviation speech analysis to improve flight safety

Importance. Ensuring flight safety is a priority for air transport. Human error, particularly communication failures in the pilot-controller system, is a critical risk factor. Even with the use of strictly regulated ICAO phraseology, verbal interactions are subject to deviations caused by stress, cognitive load, or language barriers, necessitating the development of new methods for analyzing and monitoring aviation discourse.

Research Methods. The study is based on a comprehensive linguistic approach that examines aviation communication at the lexical (deviations from standards), prosodic (tone, tempo, and pause analysis), and pragmatic (speech act analysis) levels. This approach allows for the development of a theoretical and methodological foundation for the application of machine learning and computational linguistics.

Results and Discussion. Drawing on recommendations of scientific and methodological literature, this paper analyzes the potential of modern machine learning (ML) methods and models to address human-factor-related flight safety issues. In particular, the paper examines the specifics of using automatic speech recognition (ASR), topic modeling (LDA, BERTopic), and classification architectures in this subject area: transformer (BERT), hybrid, and classical ML models based on embeddings. It is demonstrated that modern algorithms are capable of detecting not only overt protocol violations but also implicit stress markers (changes in pitch) and pragmatic mismatches (discrepancies between intention and perception). A comparative specifics analysis of using machine learning classification models to address aviation safety issues is conducted. A retrospective example of the Avianca Flight 052 disaster is used to demonstrate how multimodal ML analysis could have proactively identified a developing critical situation based on linguistic and acoustic anomalies. The article describes the development prospects for this area, related to the creation of integrated intelligent systems.

Conclusions. The symbiosis of linguistic ontology and the modern capabilities of machine learning methods creates a new paradigm for proactive aviation safety. This enables a transition from post-factum incident analysis to the creation of intelligent decision support systems for air traffic controllers, objective assessment of pilot linguistic proficiency, and the identification of latent risks in large text datasets.

1. Zhao W. (2023). A corpus-based study on aviation English from the perspective of systemic functional linguistics. Discourse & Communication, vol. 17, no. 5, pp. 630-661. https://doi.org/10.1177/17504813231171644, https://elibrary.ru/hzgvlx

2. Zhao W. (2024). A corpus-based multi-dimensional analysis of the linguistic features of aviation English. English for Specific Purposes, vol. 76, pp. 57-73. https://doi.org/10.1016/j.esp.2024.05.004

3. Nugraha D., Mayuni I., Miftahulkhairah A. (2023). Sreating an aviation English lesson: a corpus-informed material design. Journal of Teaching English for Specific and Academic Purposes, vol. 11, pp. 23-40.

4. Ishihara N., Prado M.C.D.A. (2021). The negotiation of meaning in aviation English as a lingua franca: A corpus‐informed discursive approach. The Modern Language Journal, vol. 105 (3), pp. 639-654. http://doi.org/10.1111/modl.12718, https://elibrary.ru/hzgvlx

5. Dinçer R., Dinçer N., Guksu O. (2023). An interactive conversation with a chatbot: Does ChatGPT know standard phraseology in aviation English? The Literacy Trek, vol. 9, pp. 24-41. https://doi.org/10.47216/literacytrek.1382543, https://elibrary.ru/facusm

6. Yuniar D.S., Febiyanti H., Nugraha M.E., Ramadhani S. (2024). Design of Android-based interactive e-module teaching media for aviation radiotelephony course. Didaktika: Jurnal Kependidikan, vol. 13, no. 001.

7. Dusenbury M., Bjerke E. (2013). Predictive power of English testing: Training international flight students. Journal of Aviation/Aerospace Education & Research, vol. 23, pp. 13-22. https://doi.org/10.15394/jaaer.2013.1601

8. Artamonova E.V., Galiakhmetova A.T. (2015). Monitoring and evaluation of the quality of vocational education: domestic and foreign experience. Professional’noe obrazovanie v Rossii i za rubezhom = Professional Education in Russia and Abroad, no. 1 (17), pp. 68-71. (In Russ.) https://elibrary.ru/ruwbzy

9. Lebedeva N.A., Kudryakov S.A. (2017). Language training for aviation professionals under the ICAO NGAP (next generation of aviation professionals) concept. Teoriya i praktika sovremennoi nauki = Theory and Practice of Modern Science, no. 2 (20), pp. 805-814. (In Russ.) https://elibrary.ru/yiarnh

10. Robertson F.A. (1995). Airspeak: Radiotelephony Communication for Pilots. London: Phoenix ELT Publ., 219 p.

11. Mitrofanova O.A., Atugodage M.M. (2023). Dynamic topic modelling of the Russian legal text corpus. Terra Linguistica, vol. 14, no. 1, pp. 70-87. (In Russ.) https://doi.org/10.18721/JHSS.14107, https://elibrary.ru/qlegta

12. Saifutdinov R.A., Belogrudova D.Yu., Bulat R.S. (2023). Aircraft accident prediction model using artificial intelligence. Vestnik Ul’yanovskogo uosudarstvennogo tekhnicheskogo universiteta = Bulletin of Ulyanovsk State Technical University, no. 4, pp. 51-55. (In Russ.) https://doi.org/10.61527/1684-7016-2023-4-51-55, https://elibrary.ru/kaknyr

13. Lebedeva N.A. (2017). Perspectives of introducing m-learning to universities. Vestnik Kostromskogo gosudarstvennogo universiteta. Seriya: Pedagogika. Psikhologiya. Sotsiokinetika = Vestnik of Kostroma State University. Series: Pedagogy. Psychology. Sociokinetics, no. 3, pp. 6-10. (In Russ.) https://elibrary.ru/zucjgh