Advanced AI-Driven Statistical Query Systems: A Voicebot for Seamless Demographic Data Access
摘要
This work explores the design and implementation of a voicebot to access official statistical data, in particular demographic data, using recent advances in generative AI such as Large Language Models (LLMs) and Retrieval Augmented Generation (RAG). The voicebot allows users to interact with data through natural language voice queries, providing accurate and contextual responses by combining dynamic external data with LLMs. The voicebot is designed to be particularly inclusive, benefiting people with motor disabilities (e.g. limited hand or arm mobility), visual impairments (e.g. blindness or low vision) and cognitive disabilities (e.g. learning disabilities or challenges in processing complex information). By enabling independent interaction through voice commands, it empowers users to access and engage with data autonomously. In addition, the system supports researchers, professionals, public administrators and policy makers who need quick access to demographic data for analysis and decision-making. It also supports teachers and students in educational contexts where accessibility is crucial, and provides an easier navigation option for every users who may find traditional interfaces challenging. The aim of this project is to bridge the gap between complex statistical data and a wider, more inclusive audience, fostering statistical literacy and contributing to a more informed and engaged society.