This position paper presents an application concept adapting Retrieval Augmented Generation (RAG) for geographic contexts, and combining it with Text-to-Speech (TTS) technology to automate the creation of audio narrations for locative media experiences. By minimizing the costs for research and media production, it addresses a critical barrier for archives to remediating their content and catering for audiences who are not able to, or not inclined to, read written materials. The application concept is contextualized in efforts to create alternative media representations that convey content in-situ to increase its relevance. It is illustrated with a prototype application aiming to engage people with their local history through audio narrations fictionalizing archive materials related to their location. A discussion of technical and user interface aspects offer insights into design considerations informing the prototype development.