sta - simply tag audio

A simple project to embed title and artist metadata into MP3 files - nothing fancy ♫

The frontend is built with Angular, while the backend uses Hono. The backend’s sole purpose is to send a request with uploaded file names to GPT-4o mini to identify the artist and song title as accurately as possible.

All metadata modifications for the title and artist are done locally in the browser using mp3tag.js, ensuring that no files are ever sent to the server.

Requirements

OpenAI API Key (needed for GPT-4o mini)

Installation

git clone https://github.com/devfake/sta

Backend

# Navigate to the backend directory
cd sta/backend
# Install dependencies
npm install
# Configure environment variables
cp .env.example .env
# Add your OpenAI API key to the .env file
OPENAI_API_KEY=YOUR_API_KEY
# Start the backend server
npm run dev

The backend will be available at http://localhost:3000.

Frontend

# Navigate to the frontend directory
cd sta/frontend
# Install dependencies
npm install
# Start the Angular frontend
npm start

The frontend will be available at http://localhost:4200.

Development

• Start both the Hono backend and the Angular frontend.
• The Angular frontend uses a proxy for backend calls:
  • Any API request to /api from the frontend is automatically forwarded to http://localhost:3000/api.

Production

After running the npm run build script for the Angular frontend, the production files will be placed at /backend/public. The Hono backend is configured to serve these static files at http://localhost:3000.

Rate Limiter

To protect the API from overuse, a rate limiter is enabled by default. This limits the number of API requests that can be made per hour.

You can adjust the limit by editing the RATE_LIMIT_PER_HOUR value in your .env file.