АUTOMATIC SPEECH RECOGNITION AND ASSESSMENT OF ITS APPLICABILITY IN AVIATION AT THE PRESENT STAGE

Published 2025-09-30
SOCIAL SCIENCES AND HUMANITIES Vol. 81 No. 3 (2025)
Том 81 №3 (2025)
Authors:
  • SUYUNBAEVA A.ZH.
  • SHELESTYUK E.V.
PDF (Russian)

The progress of computer technologies opens up opportunities for automatic systematization, processing and use of corpus texts, that is, the process of language recognition at the level of natural communicative speech, using special programs. The article is devoted to the use of modern technologies to reduce misunderstandings between pilots and air traffic controllers. Researchers have been studying and applying this technology for several decades. The introduction of automatic speech recognition (ASR) and machine translation technologies can significantly reduce such problems, improving safety and preventing accidents and incidents. The article presents a draft experiment, which is a comparison of the effectiveness of two popular programs for recognizing oral speech - ChatGPT (https://chatgptchatapp.com) and Speech2Text (https://speech2text.ru) - when processing texts in three languages: Russian, English and Kazakh. The study covered different types of oral speech: formalized text dictation, legal text, native language story, and natural (spontaneous) dialogue between the two participants. The analysis was carried out both in terms of transcription accuracy and taking into account the peculiarities of reproducing the structure of the text, transmitting proper names, preserving pauses and dividing speech into speakers. The body includes not only audio recordings of «pure» spoken speech, but also telephone speech, which gives the program versatility. A special complex technique for transcribing telephone conversations has been developed, since such speech is often indistinct, may contain information in other languages, have a pronounced accent, contain various speech and non-speech noises (various voice tones, sounds of preparing the speaker for the next thought, unclear words, superimposition of speech of several speakers, etc.). This system converts the original audio file into a machine-readable format. That is, it does not require additional components and transcriptions. This system is a variant of modern speech recognition technologies using an integrated model. The experimental development presented uses speech recognition technology to understand voice commands, its advantage over similar electronic applications is that it can control the smooth flow of information and its processing in extreme situations, and also has the ability to accurately reproduce natural dialogues.

SUYUNBAEVA A.ZH.

Candidate of philological sciences, associate professor, Military Institute of the Air Defense Forces, Aktobe, Kazakhstan

Е-mail: аltin_suenbaeva@mail.ru, https://orcid.org/0000-0002-0676-2028

SHELESTYUK E.V.

Doctor of Philology, Professor, Chelyabinsk State University, Chelyabinsk, Russian Federation

Е-mail: shelestiuk@yandex.ru, https://orcid.org/0000-0003-4254-4439

  1. . lderson J. C. The politics of aviation English testing. Language Assessment Quarterly, 2011 8(4), 386-403.
  2. Automatic Speech Recognition for ATC. Sesar Digital Academy: 11/05/2021
  3. Barshi I., Farris C. Misunderstandings in ATC communication. Farnham, UK: Ashgate Publishing Ltd, 2013
  4. Covelli J. M., Rolland J. P., Proctor M., Kincaid J. P., Hancock P. A., Field of view effects on pilot performance in flight. International Journal of Aviation Psychology, 2010, 20 (2), 197-219.
  5. Eurofighter Typhoon, Technology: Cockpit. 2014 URL: http://typhoon.starstreak.net/Eurofighter/cockpit.html. Last accessed: 29 March 2014.
  6. IATA, Pilots and air traffic controllers phraseology study. Montreal: International Air Transport Association, 2011
  7. ICAO, Manual of radiotelephony Doc 9432 AN/925. Montreal: International Civil Aviation Organization, 2007
  8. Jurafsky D., Martin J. H., Speech and language processing: An introduction to natural language processing, computational linguistics, and speech recognition. 2nd international edition. Upper Saddle River, NJ: Pearson Prentice Hall, 2009
  9. LDACS – The future of ATC communications. Canso. 13/01/2023 URL: https://canso.org/ldacs-the-future-of-atc-communications/#
  10. Шелестюк Е.В., Суюнбаева А.Ж., Торайғыров университетінің Хабаршысы, ISSN 2710-3439 Гуманитарлық сериясы. № 4, 2024.
ASR-MT, formal dictation, automatic speech recognition, cognitive load, identity, pipeline solutions

How to Cite

АUTOMATIC SPEECH RECOGNITION AND ASSESSMENT OF ITS APPLICABILITY IN AVIATION AT THE PRESENT STAGE. (2025). Scientific Journal "Bulletin of the K. Zhubanov Aktobe Regional University", 81(3), 171-184. https://doi.org/10.70239/