SumTool: A hybrid pipeline approach to summarise various resources

With the exponential increase in digital data, summarizing various resources has become a necessity to extract useful insights quickly and efficiently. Traditional summarization techniques have primarily focused on text documents, but with the growing popularity of multimedia content, there is a need to develop approaches that can handle different types of resources. Through this research, we introduce a hybrid pipeline approach which can summarize a variety of resources, including images, text, PDFs, and YouTube videos. The proposed approach combines different techniques, including machine learning, natural language processing, and audio analysis, to extract relevant information and generate a concise summary. We evaluate the effectiveness of our pipeline on a dataset containing various resources and show that it outperforms existing summarization techniques with a ROGUE(Recall-Oriented Under-study for Gisting Evaluation) score of 0.8 as compared to extractive (0.5) and abstractive(0.3).