Integration of GPT and holographic rendering for interactive virtual characters in intangible cultural heritage preservation

Advancements in computer technology, particularly Generative Pre-trained Transformers (GPT) and holographic rendering, have enabled the development of innovative interactive systems with broad applications in smart cities and cultural heritage preservation. This study proposes a framework integrating GPT with multimodal fusion and real-time holographic rendering to construct virtual characters capable of delivering dynamic narratives. A fine-tuned GPT model, trained on a dataset of over 1.2 million tokens, ensures linguistic accuracy and contextual fidelity, while the multimodal pipeline synchronizes text, audio, and visual elements to create lifelike holographic avatars. Experimental results demonstrate the system s effectiveness, achieving a BLEU score of 92.4\% for narrative accuracy and an average frame latency of 24 milliseconds, meeting the requirements for real-time interaction. Key innovations include the seamless integration of AI-driven semantic generation with immersive visualization techniques, providing a scalable, adaptive solution for interactive storytelling. Moreover, the holographic rendering system, leveraging advanced optical simulation techniques such as Bidirectional Reflectance Distribution Functions (BRDF), enables the realistic visualization of objects and environments, making it highly suitable for applications in urban digital twins and immersive public spaces within smart cities. This research highlights the transformative potential of computational technologies in enhancing audience engagement, promoting cultural preservation, and advancing smart city infrastructures. By bridging the gap between language processing and holographic rendering, this study establishes a foundation for next-generation virtual experiences, showcasing the potential of combining AI and optical visualization to redefine interactive systems for diverse applications.