Multilingual AI voice assistant for campus navigation and announcements

Comprehensive navigation and information access remain significant issues in vast educational campuses with linguistic diversities. This re- search proposes an innovative conversational AI system that integrates automatic speech recognition in three languages, semantic interpretation, retrieval- augmented knowledge synthesis, graph optimization, and gesture recognition to provide unified information access and navigation services in the educational campuses. The proposed system has been evaluated through systematic evaluation with 2,500 authentic user interactions via mobile apps, web portals, and kiosks. The quantitative results show that the proposed system has achieved 92.3% accuracy in intent classification, 89.2% accuracy in route finding, and sub-2-second response time with complete fidelity to the data sources under various environmental conditions, proving its superiority over traditional web portal-based information access systems.