Generative Artificial Intelligence for Effective Miner Training

This work aims to explore the use of virtual speakers generated by artificial intelligence (AI) to enhance multilingual e-learning experiences. The theoretical framework is based on the growing demand for accessible and effective e-learning platforms that overcome language barriers. Traditionally, creating audio and video content for e-learning is expensive and complex, especially in multilingual contexts. The integration of AI offers personalized and adaptive solutions, improving accessibility and student engagement. The research was conducted in the framework of the European ERASMUS+ project DigiRescueMe, which aims to develop e-learning modules to increase the knowledge and awareness of miners, rescuers and mining engineers on topics related to rescue, risk assessment and mental well-being. The project mainly targets people from Türkiye, Poland, Portugal and Italy, who often do not have a deep knowledge of English or a technical vocabulary in their own language. For this reason, a multilingual approach is essential to effectively engage users. The experiment involved 147 participants from different educational and professional backgrounds. The article highlights the potential of AI-generated virtual speakers to improve the accessibility and effectiveness of multilingual e-learning. Despite a certain preference for human speakers in some contexts, AI offers significant advantages in terms of customization, scalability, and cost reduction. It is crucial to address AI limitations, such as data quality and algorithmic biases, and integrate AI with human instruction to create more balanced and effective learning environments.