Chrome Extension :Video Summarizer

Authors: Asst.Prof.Kedar Pinniboniya, Rudresh Choubey, Mayank Kumar, IshitaSanya, Nishi Priyanka Hembrom

Abstract: With the exponential growth of video content on platforms like YouTube, accessing key information efficiently has become a significant challenge. This paper proposes the design and implementation of a Chrome Extension integrated with a Flask-based API that automatically extracts YouTube video transcripts and generates concise summaries using Natural Language Processing (NLP). The system leverages AI-based summarization models to provide users with quick insights, thereby improving accessibility and saving time. The proposed framework operates in three stages: 1. Automated transcript retrieval using the YouTube Data API, 2. Preprocessing and re- finement of textual content through NLP pipelines, 3. Generation of abstractive or extractive summaries using transformer-based language models. The summarized outputs are then presented to users via a lightweight Chrome Extension interface, allowing seamless access without leaving the YouTube platform. This approach offers significant advantages for diverse user groups including students, researchers, and professionals who rely on video lectures, tutorials, or informational content for learning and decision making. By reducing information overload and minimizing the time required to grasp core ideas, the tool enhances productivity and accessibility. Experimental evaluations indicate that the generated summaries achieve high levels of relevance and coherence compared to raw transcripts, validating the utility of the system. Ultimately, this work demonstrates how AI-driven summarization can transform video-based learning and information retrieval into a more efficient and user-friendly experience.

DOI: https://doi.org/10.5281/zenodo.17097304