Abstract: The vast amount of information on YouTube is often embedded in lengthy videos, making it time-consuming for users to extract key insights. To address this inefficiency, this project introduces a Smart YouTube Video Summarizer that generates quick and meaningful summaries. The system uses the YouTube API to fetch video details and transcript, then processes the content using Natural Language Processing (NLP) and the LLaMA3 model to create concise summaries. Users can paste a video URL, view the transcript, and get the summary instantly in a Streamlit web application. The project reduces the time required to watch long videos, provides an option to view the transcript with timestamps, and supports multiple languages. This system is efficient, user-friendly, cost-effective, and helpful for students, researchers, and general users who need information quickly.

Keywords: YouTube Summarizer, Transcript Extraction, Natural Language Processing, Abstractive Summarization, LLaMA3, Streamlit.

Downloads: | DOI: 10.17148/IARJSET.2025.121018