Bridging Knowledge and Technology: Constructing a Knowledge-Driven Repository with 3D CNN for Interpreting Education

Against the backdrop of digital humanities, artificial intelligence (AI) has been increasingly integrated with language education, particularly interpreting training. Although the academic world has reached a consensus about the importance of constructing a shared and intelligent repository for interpreting education by collecting speech videos, few have proposed workable methods of efficiently curating these collected resources. The current study presents a comprehensive exploration of synergizing a 3D Convolutional Neural Networks (CNN) model and a knowledge graph to enhance the management and usage of the collated videos. Drawing upon widely used textbooks for interpreting training, the current study demonstrates the construction of a knowledge graph (KG) of interpreting knowledge and skills, and utilizes it to navigate the video-sorting process supported by a 3D CNN model. By infusing the video repository with AI, this endeavor may provide students with personalized learning experiences, thus stimulating their learning interests in interpreting. The implications extend to instructors and material designers, providing them with insights about the overview of interpreting education and the roadmap of organizing resources. Lastly, this study is a cross-disciplinary exploration that bridges interpreting training and AI.