The rapid advancement of Artificial Intelligence (AI) and Augmented Reality (AR) has transformed business communication, particularly in startup pitching. Traditional presentation methods rely on static tools such as PowerPoint, which lack interactivity and fail to engage investors effectively. Additionally, generating structured pitches from unorganized business documents is time-consuming, and existing AI systems often produce inaccurate responses due to lack of contextual grounding. This paper proposes an AI-based AR-driven personal avatar system that leverages Retrieval-Augmented Generation (RAG) and Large Language Models (LLMs) to generate structured elevator pitches and provide accurate responses to investor queries. The system integrates Text-to Speech (TTS) technology and a Unity-based 3D avatar to deliver immersive presentations. Experimental results demonstrate improved accuracy, reduced hallucination, and enhanced user engagement. The proposed system provides an intelligent and Anurag Nagapure Department of Computer Engineering MMCOE, Pune, India. combines information retrieval with generative models. RAG enhances the factual accuracy of responses by retrieving relevant information from external knowledge sources and using it as context for generation 5. Recent studies have shown that RAG significantly improves performance in question-answering systems and reduces hallucination in conversational AI applications 6, 7. Furthermore, advancements in vector databases and embedding techniques have enabled efficient semantic search and retrieval, making RAG-based systems more scalable and reliable 8. interactive solution for modern business pitching.
Kalhapure et al. (Fri,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: