Generative AI in mixed reality for cabin crew training
The current training method for cabin crew in commercial airlines faces inefficiency and scalability challenges due to fixed mock-up capacities and diverse plane models requirements, prompting the need for innovative solutions.
Virtual reality (VR) emerges as a reliable alternative, offering mobility, flexibility, and cost-effectiveness. We leverage the power of a Large Language Model (LLM) to analyse and generate responses as the virtual passenger. This allows us to generate a more unpredictable situation yet able to provide objective feedback to the trainees through tracking of various data. Speech-to-Text (STT) and Text-to-Speech (TTS) algorithms are also utilised to analyse and generate speech that is more closely aligned with a Singaporean setting. This project explores the integration of these technologies to create a convincing training environment with accurate and natural interactions, addressing the pressing demand for scalable and effective cabin crew training solutions.
Project Team
Students:
- Chen Qing Hua (Mechanical Engineering, Class of 2024)
- Chen Yuhan (Electrical Engineering, Class of 2024)
- Raymond Bala (Electrical Engineering, Class of 2024)
Supervisor:
- A/Prof Khoo Eng Tat (etkhoo@nus.edu.sg)