Update README.md
Browse files
README.md
CHANGED
|
@@ -1,89 +0,0 @@
|
|
| 1 |
-
|
| 2 |
-
# VideoChad🗿 : RAG based Youtube Video Conversational Chat-Bot 📤📺
|
| 3 |
-
|
| 4 |
-
### *Got Bored watching Long Youtube Videos ? Here's a Full Stack App that*
|
| 5 |
-
- ⭐ **Generates Smart Summary** ⭐
|
| 6 |
-
- ↪ **Provides BackLinks (Reference) to the Video (No hallucination)** ↪
|
| 7 |
-
- 🗣 **(ChatBot) Enables you to have Conversation with Video** 🗣
|
| 8 |
-
- 🧠**Generates MindMap** 🧠
|
| 9 |
-
|
| 10 |
-
## Demo
|
| 11 |
-
|
| 12 |
-
[](https://www.youtube.com/watch?v=_fflcGaQjBM)
|
| 13 |
-
|
| 14 |
-
|
| 15 |
-
## Features
|
| 16 |
-
|
| 17 |
-
- **Automated Video Summarization**: The application generates concise summaries of video content, allowing users to grasp key concepts efficiently and identify areas requiring deeper focus.
|
| 18 |
-
- **Real-time Chat Interaction**: Users can engage in conversation with the video content, fostering a deeper understanding of the subject matter by asking questions and receiving instant responses.
|
| 19 |
-
- **Video Backlinking**: The application incorporates a backlinking feature that enables users to seek relevant timestamps in the video player by clicking on provided reference links.
|
| 20 |
-
- **MindMap**: Generates a interactive mindmap using the important keywords from the video content's essence!
|
| 21 |
-
- **Transcript Download**: Users can download a text file containing the transcript of the processed video for future reference.
|
| 22 |
-
|
| 23 |
-
|
| 24 |
-
## Technologies Used
|
| 25 |
-
|
| 26 |
-
- **Flask**: A lightweight Python web framework used for building the backend API.
|
| 27 |
-
- **React**: A JavaScript library for building the user interface on the frontend.
|
| 28 |
-
- **Large Language Models (LLMs)**: Specifically, the OpenAI ChatGPT 3.5 (gpt-3.5-turbo) model is employed for generating contextual responses.
|
| 29 |
-
- **Retrieval-Augmented Generation (RAG)**: This approach combines a retriever and a language model, allowing for efficient retrieval of relevant information from the video transcript.
|
| 30 |
-
- **LangChain**: A framework for building applications with large language models, facilitating the integration of the RAG approach.
|
| 31 |
-
- **Vector Database (Chroma)**: A vector database used for storing and efficiently searching the embeddings of the video transcript.
|
| 32 |
-
- **OpenAI Embeddings API**: Utilized for converting textual data into high-dimensional vector representations.
|
| 33 |
-
- **YouTube API**: Employed for fetching video transcripts and metadata.
|
| 34 |
-
|
| 35 |
-
## Getting Started
|
| 36 |
-
|
| 37 |
-
To get started with this application, follow these steps:
|
| 38 |
-
|
| 39 |
-
1. Clone the repository:
|
| 40 |
-
```
|
| 41 |
-
git clone https://github.com/foolmalhar/VideoChad.git
|
| 42 |
-
```
|
| 43 |
-
|
| 44 |
-
2. Install the required dependencies:
|
| 45 |
-
```
|
| 46 |
-
cd VideoChad
|
| 47 |
-
pip install -r requirements.txt
|
| 48 |
-
```
|
| 49 |
-
|
| 50 |
-
3. Set up the necessary environment variables:
|
| 51 |
-
- `OPENAI_API_KEY`: Your OpenAI API key for accessing the language models. [OpenAi Platform](https://platform.openai.com/account/api-keys) | [Account Setup](https://platform.openai.com/docs/quickstart/account-setup)
|
| 52 |
-
|
| 53 |
-
4. Start the Flask backend:
|
| 54 |
-
```
|
| 55 |
-
python app.py
|
| 56 |
-
```
|
| 57 |
-
|
| 58 |
-
5. In a separate terminal, start the React frontend: (optional)
|
| 59 |
-
```
|
| 60 |
-
cd VideoChad
|
| 61 |
-
npm install
|
| 62 |
-
npm run dev
|
| 63 |
-
```
|
| 64 |
-
|
| 65 |
-
6. Access the application in your web browser at `http://localhost:5000`.
|
| 66 |
-
( if you don't use static pre-built files and are running node on the VideoChad frontend, then port might be :3000, check terminal )
|
| 67 |
-
|
| 68 |
-
## Usage
|
| 69 |
-
|
| 70 |
-
1. Enter a valid YouTube video link in the provided input field. (Link must have English Transcript available on Youtube )
|
| 71 |
-
2. The application will fetch the video transcript and generate a summary.
|
| 72 |
-
3. Interact with the video content by asking questions in the chat interface.
|
| 73 |
-
4. Click on the provided reference links to seek relevant timestamps in the video player.
|
| 74 |
-
5. explore !
|
| 75 |
-
|
| 76 |
-
## Contributing
|
| 77 |
-
|
| 78 |
-
Contributions to this project are welcome! If you encounter any issues or have suggestions for improvements, please open an issue or submit a pull request.
|
| 79 |
-
|
| 80 |
-
## License
|
| 81 |
-
|
| 82 |
-
This project is licensed under the [MIT License](LICENSE).
|
| 83 |
-
|
| 84 |
-
## Acknowledgments
|
| 85 |
-
|
| 86 |
-
- [DeepLearning.ai Short Course](https://www.deeplearning.ai/short-courses/langchain-chat-with-your-data/) Understand RAG with Langchain and Chromadb
|
| 87 |
-
- [LangChain](https://www.langchain.com/) The Tool!
|
| 88 |
-
- [OpenAI](https://openai.com/) for their powerful language models and APIs.
|
| 89 |
-
- [Chroma](https://www.trychroma.com/) for their vector database solution.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|