[DATN] AI-Based Application for Recording, Analyzing, and Evaluating Technical Diagnostic Processes

This project develops an AI-based application for analyzing technical diagnostic conversations. The system is designed to improve the efficiency of managing and processing customer interactions by integrating modern web technologies with Speech-to-Text functionality. The application follows a client–server architecture, using NextJS for the frontend, NestJS for the backend, and MySQL for data storage. A key feature of the system is the ability to convert uploaded audio files into text automatically, enabling structured storage and easier retrieval of conversation data. In addition, role-based access control (RBAC) is implemented to support different user roles, including Administrator, Staff, and Customer. The system was tested across core functionalities such as authentication, audio processing, and conversation management, showing reliable performance. However, transcription accuracy depends on audio quality, and the system does not yet support real-time processing. Overall, the project demonstrates the practical application of Artificial Intelligence in conversation management systems and provides a foundation for future improvements in performance and analytical capabilities.