Aditya-y9 / WhyT

Introducing WhyT 🚀 – your ultimate tool for seamless YouTube-to-PDF note transformation! 📝 Harnessing the power of Machine Learning, WhyT automagically captures screenshots and crafts concise text summaries, revolutionizing your note-taking game. 🤖 Say farewell to the drudgery of manual note-making and embrace efficient content consumption.

📱Screenshots

WHYT -The class topper's own notes taker
CoC Inheritance 2023 || HackElite

---

Table of Contents

• Description
• Links
• Tech Stack
• Progress
• Future Scope
• Applications
• Project Setup
• Usage
• Team Members
• Mentors
• Screenshots

📝Description

Our project is a comprehensive tool designed to benefit a wide range of users, including students, analysts, and anyone seeking efficient information retrieval. We have developed a YouTube video summarizer coupled with an important slides generator, packaged in the form of PDFs.

1.This feature enables users to extract key points and summaries from YouTube videos. It condenses lengthy videos into concise and easily digestible summaries.

2.Our tool automatically identifies significant slides within a presentation or lecture video. It extracts these slides and compiles them into a PDF document for easy reference.

3.Our integrated chatbot allows users to interact with the video content through questions and answers.Users can ask questions related to the video and receive responses based on the content discussed.

4.We've crafted our website with the MERN stack—MongoDB, Express.js, React, and Node.js—to seamlessly blend front-end and back-end elements. Our user interface, styled with Tailwind CSS, offers a sleek and contemporary appearance, enhancing user interaction.

🔗Links

• GitHub Repository
• Demo Video

🤖Tech-Stack

Front-end

• ReactJS

• Taiwind CSS

Back-end

• NodeJS
• ExpressJS

Database

• MongoDB

Python Libraries

transformers

transformers is a python library that exposes an API to use many well-known transformer architectures, such as BERT, RoBERTa, GPT-2 or DistilBERT, that obtain state-of-the-art results on a variety of NLP tasks like text classification, information extraction, question answering, and text generation.

moviepy

moviepy is a Python library for video editing: cutting, concatenations, title insertions, video compositing (a.k.a. non-linear editing), video processing, and creation of custom effects.

torch

torch is a machine learning library that can be used for video summarization and slide extraction.

torchvision

torchvision is a library that can be used for video summarization and slide extraction.

imagehash

imagehash is a Python library that can be used to extract slides from a video.

FPDF

FPDF is a Python library that can be used to convert slides into a PDF.

langchains

langchains is a Python library that can be used to convert text into a summary.

pytube

pytube is a Python library that can be used to extract video content from YouTube.

flask

flask is a Python library that can be used to create a web application.

📈Progress

List down all the fully implemented features in your project

• Summarization of YouTube videos (Summarizer) Our tool can extract key points and summaries from YouTube videos. It condenses lengthy videos into concise and easily digestible summaries. It is divided into 3 parts:

1. Video Download

   The user can input the URL of the video they want to summarize. The tool will then download the video from YouTube.

2. Text Extraction

   The tool will then extract the text from the video.

3. Summarization

   The tool will then summarize the text and display the summary to the user.

   • Important Slides Generation (Slides Extractor) Our tool automatically identifies significant slides within a presentation or lecture video. It extracts these slides and compiles them into a PDF document for easy reference. It is divided into 3 parts:

4. Video Download

   The user can input the URL of the video they want to summarize. The tool will then download the video from YouTube.

5. Slide Extraction

   The tool will then extract the slides from the video.

6. Slide Classification

   The tool will then classify the slides and display the PDF to the user.

   Basis of Classification

   • Slide with text
   • Slide with image
   • Slide with both text and image
   • Nth slide is not similar to the previous slide (N-1)th slide.

• Chatbot Integration Our integrated chatbot allows users to interact with the video content through questions and answers. Users can ask questions related to the video and receive responses based on the content discussed.

🔮Future Scope

• In our project, we aspire to expand functionality by not only allowing users to input playlist URLs but also providing comprehensive summaries and detailed notes for each playlist. This enhancement aims to offer users a deeper understanding of the content within the playlists, facilitating quicker navigation and comprehension
• In addition to our current objectives, we are looking to implement a feature that generates graphs for each topic within a playlist. These graphs will visually represent the relationships, connections, and frequencies of topics present in the playlist content. By visualizing the structure of topics, users will gain a clearer understanding of the playlist's thematic landscape, allowing for easier navigation and comprehension.

💸Applications

1)Educational Aid: -Students: Helps students in quickly grasping key concepts from educational videos and lectures. -Educators: Assists educators in creating concise summaries and slide decks for their teaching materials.

2)Professional Analysis and Research: -Analysts and Researchers: Provides analysts and researchers with a tool for efficiently extracting insights and important information from lengthy videos, enhancing productivity in data analysis and research.

Application of Youtube URL - To - Summary and Screenshots PDF generation

1. Students can use this tool to summarize their lectures and create PDFs of important slides.
2. Analysts can use this tool to summarize important videos and create PDFs of important slides.
3. Anyone seeking efficient information retrieval can use this tool to summarize important videos and create PDFs of important slides.

Monetization Strategy We can charge a fee for the use of our tool in the following ways:

1. We can charge a subscription fee for the use of our tool.
2. We can charge a fee for the use of our tool on a per-use basis.
3. We can charge a fee for the use of our tool on a per-video basis.

Advertising

1. We can display advertisements on our website.
2. We can display advertisements on our tool.
3. We can display advertisements on our PDFs.

🛠Project Setup

Clone the repository

git clone https://github.com/kasodeep/inheritance-project.git

Install the required packages

pip install -r requirements.txt

cd frontend
npm install

cd ..

cd backend
npm install

💻Usage

Run the following command to start the server

python app.py

Open new terminal and run the following command

cd backend
npm run dev

Open new terminal and run the following command

cd frontend
npm run dev

Congrats! You just started the project on your local machine. go to http://localhost:3000/ to see the project live.

👨‍💻Team Members

• Team Member 1 Nishit Kekane: [e-mail]
• Team Member 2 Deep Kasodariya: [e-mail]
• Team Member 3 Ayush Gulhane: [e-mail]
• Team Member 4 Aditya Yedurkar: [e-mail]

👨‍🏫Mentors

• Mentor 1 Anushree Dere: [e-mail]

• Mentor 2 Srushti: [e-mail]

  Special Thanks to Pranav Janjani:[e-mail] for helping and supporting ML team throughout the project.