Text Humanizer

A sophisticated Flask-based application that rephrases text generated by large language models (LLMs) to make it sound more natural and human-like. The tool employs advanced techniques including context-aware rephrasing, grammar enhancement, and semantic restructuring.

Features

• Real-time text humanization
• Context-aware rephrasing
• Secure API with CSRF protection
• Caching for improved performance
• Response compression
• Detailed error handling
• Logging and monitoring

Quick Start

Prerequisites

• Python 3.11 or higher
• pip (Python package manager)
• Virtual environment (recommended)

Installation

1. Clone the repository:

git clone https://github.com/yourusername/re-phrasing-tool.git
cd re-phrasing-tool

2. Create and activate a virtual environment:

python -m venv .venv
source .venv/bin/activate  # On Windows: .venv\Scripts\activate

3. Install dependencies:

pip install -r requirements.txt

4. Configure the application:

• Copy text_humanizer/config/config.example.json to text_humanizer/config/config.development.json
• Adjust settings as needed

Running the Application

1. Set the environment:

export APP_ENV=development  # On Windows: set APP_ENV=development

2. Start the server:

python -m text_humanizer.main

The application will be available at http://127.0.0.1:5000

API Documentation

REST Endpoints

POST /api/humanize

Humanizes the provided text using context-aware rephrasing.

Request Body:

{
    "text": "Text to be humanized",
    "context": "Optional context for better understanding",
    "style": "Optional style preferences"
}

Response:

{
    "status": "success",
    "humanized_text": "Rephrased human-like text",
    "confidence_score": 0.95
}

GET /api/context

Retrieves available context segments for text humanization.

Response:

{
    "status": "success",
    "contexts": [
        {
            "id": "context-1",
            "description": "Context description",
            "active": true
        }
    ]
}

Architecture

Component Overview

text_humanizer/
├── main.py           # Application entry point
├── input_processor/  # Text preprocessing
├── context_manager/  # Context handling
├── llm_client/      # LLM integration
├── providers/       # Service providers
└── utils/          # Utility functions

Key Components

1. Input Processor: Handles text preprocessing and validation
2. Context Manager: Manages context storage and retrieval
3. LLM Client: Interfaces with language models
4. Local LLM Provider: Implements local model integration
5. Config Manager: Handles configuration and hot-reloading

Data Flow

1. Request received → Input validation
2. Context retrieval and processing
3. LLM processing with context
4. Response formatting and caching
5. Compressed response delivery

Configuration

The application uses a hierarchical configuration system:

1. Default configuration (config.default.json)
2. Environment-specific configuration (config.[environment].json)
3. Environment variables (override file-based config)

Key Configuration Options

{
    "debug_mode": false,
    "secret_key": "your-secret-key",
    "csrf_enabled": true,
    "cache_settings": {
        "type": "simple",
        "timeout": 300
    },
    "model_settings": {
        "model_name": "your-model",
        "temperature": 0.7
    }
}

Deployment

Production Deployment

1. Set up a production server (e.g., Ubuntu 20.04 LTS)
2. Install Python 3.11 and required packages
3. Configure a reverse proxy (nginx recommended)
4. Set up SSL/TLS certificates
5. Configure environment variables:

   export APP_ENV=production
   export SECRET_KEY=your-secure-key

6. Use a process manager (e.g., supervisord)

Docker Deployment

FROM python:3.11-slim
WORKDIR /app
COPY . .
RUN pip install -r requirements.txt
CMD ["python", "-m", "text_humanizer.main"]

Build and run:

docker build -t text-humanizer .
docker run -p 5000:5000 text-humanizer

Contributing

1. Fork the repository
2. Create a feature branch
3. Commit your changes
4. Push to the branch
5. Create a Pull Request

License

This project is licensed under the MIT License - see the LICENSE file for details.

Support

For support, please open an issue in the GitHub repository or contact the maintainers.