Integration of artificial intelligence in the teaching of technical english for communications personnel of technical english for communications personnel of the Ecuadorian Air Force

Abstract

The integration of artificial intelligence (AI) into the teaching of technical English for Ecuadorian Air Force communications personnel represents a vital advance in optimizing language proficiency in critical aeronautical contexts. This study focuses on evaluating the effectiveness of adaptive intelligent systems, which adjust training content and difficulty according to individual needs, with a special emphasis on high-precision communication operations. The growing demand for solutions based on machine learning and natural language processing has driven the development of advanced technologies, including realistic simulations and automatic speech recognition (ASR), improving the fidelity and applicability of training in controlled environments. The study employs a mixed methodology, combining advanced statistical models such as mixed ANOVA, multinomial logistic regression, and Bayesian analysis to evaluate the influence of variables such as learning modality, training frequency, motivation, cognitive load, and level of adaptive AI. The results indicate that, while the learning modality (face-to-face, virtual, or mixed) does not produce statistically significant differences, training frequency has a marginal positive impact on final performance. Furthermore, high individual variability is identified, influenced by motivational factors and prior experience. This evidence validates the importance of frequent and consistent training, regardless of the instructional format. This work contributes to closing gaps in the rigorous evaluation of intelligent systems in specialized military contexts and offers recommendations for flexible and personalized approaches to technical English training. The need for future research with real-life operational data and validation of individualized predictive models is highlighted.