Story Generator with Text-to-Speech

A Flask web application that generates stories and provides text-to-speech functionality using multiple TTS services.

My son likes to listen to story before he sleeps, especially with his mom's voice. So I created this app with standard google tts service first and implement voice clone function with f5 tts.

Features

• Story generation with AI
• Multiple TTS services:
  • Google Text-to-Speech (gTTS)
  • F5 Text-to-Speech
• User authentication system
• Voice profile management
• Audio streaming
• Database integration for story storage

Installation

1. Clone the repository:

git clone https://github.com/yourusername/story-generator.git
cd story-generator

2. Install dependencies:

pip install -r requirements.txt

3. Set up configuration:

cp config_template.py config.py

Edit config.py with your API keys and settings

Running the Application

Start the development server:

python run.py

The application will be available at http://localhost:5000

Start the f5 tts server:

python f5tts_api.py

You need to install f5-tts first. For the installation of f5 tts, you can refer to the official f5 tts library

Project Structure

story-generator/
├── app/                  # Main application package
│   ├── auth/             # Authentication module
│   ├── db/               # Database operations
│   ├── story/            # Story generation logic
│   ├── tts/              # Text-to-speech services
│   ├── utils/            # Utility functions
│   └── static/           # Static files (CSS, JS, images)
├── scripts/              # Database management scripts
├── voice_profiles/       # Voice profile assets
├── config_template.py    # Configuration template
├── requirements.txt      # Python dependencies
└── run.py                # Application entry point

Configuration

Edit config.py to set up:

• Database path
• API keys
• TTS service preferences
• Application settings

Contributing

Pull requests are welcome. For major changes, please open an issue first to discuss what you would like to change.

I have very limited knowledge about the frontend, so almost everything in the frontend is generated by cursor. The backend is also generated by cursor first and I review/modify/debug it. If anyone can help improve the frontend, I will appreciate it.

License

MIT