I built AI Web Platform.

Hi everyone! I’d like to share my project - AI Web Platform.

It’s a full-stack platform for creating AI-generated video projects with a scene-based workflow. Users can create a project, generate or edit scripts, split them into scenes, generate videos, images with different AI models, manage a timeline, and add music, speech, or sound effects.

The platform is built with Vue, FastAPI, SQLite, WebSockets, and modular AI model integrations, including video, image, music, TTS, and sound-generation APIs. The main idea is to make AI video production more structured, editable, and convenient for creative workflows.

Happy to share more details or discuss possible use cases.

Here's the repository.

thehemen / ai_platform

Video Generation Platform

A full-stack application for AI-powered video generation with FastAPI backend and Vue.js frontend.

Features

• User authentication and personal workspaces
• OpenAI-powered script generation
• Multi-scene video projects with editable timelines
• Multiple video generation models (Fal.ai Seedance v1.5 Pro)
• WebSocket real-time updates
• Video preview and export functionality

Architecture

• Backend: FastAPI (Python)

  • RESTful API with JWT authentication
  • WebSocket support for real-time updates
  • SQLite database for data persistence
  • Modular video API integration system
  • OpenAI integration for script generation

• Frontend: Vue 3 + Vite

  • Pinia for state management
  • Vue Router for navigation
  • Axios for API communication
  • WebSocket integration

Setup

Prerequisites

• Docker and Docker Compose
• OpenAI API key
• Fal.ai API key

Configuration

1. Copy the environment file template:

   cp .env.example environment

2. Edit environment file and add your API keys:

   OPENAI_API_KEY=your_openai_api_key_here
   FAL_API_KEY=your_fal_api_key_here
   SECRET_KEY=your-secret-key-change-this
   

Running with Docker

1. Build and start the containers:

   docker-compose up --build

2. Access the application:

   • Frontend: http://localhost:5173
   • Backend API: http://localhost:8001
   • API Documentation…

View on GitHub

Good luck!