The development of voice-assisted chatbot for healthcare institutions using transformers-based techniques

Lee, Zong Hao (2025) The development of voice-assisted chatbot for healthcare institutions using transformers-based techniques. Final Year Project, UTAR.

Preview

PDF Download (4Mb) | Preview

Abstract

This research focuses on developing a voice-assisted chatbot tailored for Traditional Chinese Medicine (TCM), utilizing advanced transformer-based AI models and generative techniques. The chatbot aims to address accessibility challenges in healthcare services, particularly for elderly, disabled, or literacy-challenged individuals. By enabling voice input and output, it ensures inclusivity and broader access to essential healthcare information. The chatbot's core features include speech recognition and Natural Language Processing (NLP), allowing it to understand various dialects and accents, a critical need in multicultural regions like Malaysia. Leveraging large language models, it provides human-like responses and personalized recommendations based on TCM principles, including herbal remedies, dietary advice, and lifestyle suggestions. It is designed to function effectively even in noisy environments and to understand accented English, ensuring accurate communication across diverse linguistic backgrounds. Additionally, the chatbot bridges the gap in TCM-specific medical knowledge by using specialized datasets to deliver detailed insights into treatments and principles. It serves as an educational resource for patients and practitioners, continually improving its responses through dynamic learning from user interactions. This voice-assisted TCM chatbot significantly enhances healthcare workflows by reducing the burden on medical professionals, automating routine consultations, and providing 24/7 access to medical advice. It is especially beneficial for users in remote areas, offering timely and accurate information. By personalizing care and empowering users to manage their health proactively, this project represents a major step forward in integrating AI into healthcare, creating an inclusive and patient-centric system.

Actions (login required)