StorySpinner: AI Based Technique to Narrate the Web Links

The rapid advancement of LLM has facilitated a variety of large language model applications in natural language processing. This report details the development of StorySpinner, an application based on large language models and retrieval-augmented generation that allows yield from articles to the format of an interesting narrative. Moreover, the model addresses processing issues related to the specifics of diverse Internet re-sources, webpages, blogs, and even YouTube videos, by combining web scraping methods, text summarization models, and semantic paragraphs with the use of the extraction of material from extracted abstracts into LLM. The present research investigated the specifics of StorySpinner architecture, the used techniques of information extraction, summarization, and narrative generation, and possible applications of this technology such as educational purposes, accessibility issues, and content development. The paper also provided a discussion of ethical aspects, including bias prevention, copyright issues, and AI technology influence on society. StorySpinner was introduced as a method to fill the gap between web content and human understanding, offering a new way to access information as a story.