Youtube Timeline Generator

Description

This project is aimed at generating a video timeline description using LLMs. It allows users to generate summary and descriptive timelines for videos, making it easier to navigate through video.

Demo

Project Setup

1. Install Poetry:

   curl -sSL https://install.python-poetry.org | python3 -

2. Clone the repository

3. Navigate to the project directory

4. Create environment and Install project dependencies:

   poetry shell
   poetry install

Usage

Running the project for the first time?

1. Run the helper script for deploying the LlamaEdge API Server on this machine and starting the server:

sh start.sh

The application can be reached at http://localhost:5443.

Usage via Docker

Before you start, ensure your system meets the following minimum requirements to run the project using Docker:

• CPU: 8 cores
• Memory: 12GB
• Disk: 20GB

Building the Docker Image

First, you have to build the Docker image for the project. Navigate to the project's root directory and run the following command:

docker build -t youtube-timeline-generator:dev .

Running the Docker Container

After the image has been built, you can run the Docker container with the following command:

docker run -p 5443:5443 youtube-timeline-generator:dev

The application can be reached at http://localhost:5443.

System Diagram

Components

• UI
  • Bootstrap and jQuery
• Backend
  • APIs exposed via Flask
• LLM Agent via LamaEdge
  • LLM model - SanctumAI/Meta-Llama-3-8B-Instruct-GGUF
  • Model exposed over APIs using LlamaEdge

Features & Extensions

• User-friendly interface
• Validations on user input
• Easy copy-paste of summary and description

Extensions

• History page - showing past requests
• More unit and Integration tests
• Improve response time

License

This project is licensed under the MIT License.