Renaissance of AI-Driven Semiotic Journey Narratives Through Visual Art: Cultural Heritage and History of Islam in Patani's Talaa Gong Fabric Patterns

Abstract

This research integrates multi-model language learning models (LLMs) in Generative AI, specifically Variational Autoencoders (VAEs), to preserve and represent cultural heritage through video art. Focusing on the Patani state, now encompassing the provinces of Pattani, Yala, and Narathiwat in southern Thailand, the study explores the historical role of Islam in shaping regional cultural practices. Talaa Gong garments serve as a key medium to symbolize the area’s identity and memory. The study investigates two core aspects: the historical shifts of Islamic influence and its effects on local traditions, and the creation of modern textile patterns for women’s Talaa Gong garments, complemented by video art reflecting the social dynamics of the region. Through fieldwork, including interviews with religious leaders, artists, and residents, the research demonstrates that combining Generative AI with artistic practices enhances cross-cultural engagement and cultural preservation. The findings reveal that this AI-driven approach offers a more refined analysis of cultural evolution, contributing to semiotic journey narratives via visual art, with a focus on the heritage and history of Islam in Patani's Talaa Gong fabric.