Subtitles for Everyone (SFE)

A powerful real-time screen text translation application that instantly translates subtitles and text overlays in movies, TV shows, games, and any on-screen content using advanced OCR technology and DeepL's high-quality translation API. Break language barriers and enjoy global content without missing a beat!

Demo Video: The videos below are for promotional purposes only and may contain commercial images. Copyright belongs to the game developers. They are used for educational/demo purposes.

Youtube

Game

Features

• Real-time OCR: Captures text from selected screen areas using Tesseract OCR
• Multiple Translation Modes:
  • Grayscale threshold
  • Adaptive threshold
  • Color filtering (HSV-based)
• DeepL Integration: High-quality translations using DeepL API
• System Tray: Runs in background with system tray icon
• Customizable UI: Adjustable fonts, colors, transparency, and positioning
• Hotkey Support: Keyboard shortcuts for area selection, pause/resume, and exit
• Multi-language Support: Interface available in multiple languages
• Smart Filtering: Avoids duplicate translations and filters short text

Requirements

• Python 3.9+
• Tesseract OCR (installed and added to PATH)
• DeepL API key
• Windows OS (tested on Windows 10/11)

Python Dependencies

opencv-python
pytesseract
deepl
keyboard
pystray
Pillow
mss
ttkthemes

Installation

1. Clone the repository:

   git clone https://github.com/Enesuygurs/sfeapp.git
   cd sfeapp

2. Install Python dependencies:

   pip install -r requirements.txt

3. Install Tesseract OCR:

   • Download from: https://github.com/UB-Mannheim/tesseract/wiki
   • Install and add to system PATH
   • Test installation: tesseract --version

4. Get DeepL API Key:

   • Sign up at: https://www.deepl.com/pro-api
   • Get your API key from the dashboard

Configuration

1. First Run Setup:

   • Run python sfe.py
   • Configure Tesseract path in settings
   • Enter your DeepL API key
   • Select target language

2. Area Selection:

   • Use hotkey (default: F8) to open area selector
   • Click and drag to select the area to monitor
   • The application will capture text from this area

3. OCR Settings:

   • Grayscale Threshold: Simple binary thresholding
   • Adaptive Threshold: Dynamic thresholding for varying lighting
   • Color Filter: HSV-based color filtering for specific text colors

Usage

1. Start the Application:

   python sfe.py

2. System Tray Controls:

   • Right-click the tray icon for menu options
   • Select Area: Choose screen area to monitor (F8)
   • Pause/Resume: Toggle translation (F9)
   • Settings: Open configuration window
   • Exit: Close application (F10)

3. Translation Process:

   • Application continuously captures the selected area
   • OCR extracts text from the image
   • Text is translated using DeepL API
   • Translated text appears as an overlay on screen

Configuration Options

General Settings

• Tesseract Path: Path to tesseract.exe
• DeepL API Key: Your DeepL API authentication key
• Target Language: Language to translate to
• Interface Language: Application UI language
• Scan Interval: Time between screen captures (seconds)
• Similarity Threshold: Minimum text similarity to trigger new translation
• Min Text Length: Minimum character count for translation

OCR Settings

• Processing Mode: Grayscale, Adaptive, or Color filtering
• Threshold Value: Binary threshold for grayscale mode
• Auto Invert: Automatically detect and invert dark text on light backgrounds
• HSV Color Range: Color filtering parameters (H, S, V min/max)

Interface Settings

• Font: Family, size, style (bold, italic, underline)
• Colors: Font and background colors
• Transparency: Overlay opacity (0.1-1.0)
• Position: Vertical offset from top of screen
• Translation Lifespan: How long translations remain visible (seconds)

Hotkeys

• Select Area: Open area selection tool
• Pause/Resume: Toggle translation on/off
• Exit: Close application

Project Structure

sfeapp/
├── sfe.py                 # Main application file
├── gui.py                 # GUI components and settings window
├── config_manager.py      # Configuration management
├── ocr_tester.py          # OCR testing and preview tool
├── config.ini            # User configuration file
├── target_languages.json   # Supported target languages
├── interface_languages.json # Supported interface languages
├── lang/                 # Language files
│   ├── en.json
│   ├── tr.json
│   ├── de.json
│   ├── es.json
│   └── fr.json
├── images/               # Application icons
│   ├── icon.png
│   ├── pause_icon.png
│   └── stop.png
└── README.md

How It Works

1. Screen Capture: Uses mss library to capture selected screen area
2. OCR Processing:
   • Converts image to appropriate format (grayscale/color filtering)
   • Applies thresholding or color filtering
   • Uses Tesseract to extract text
3. Text Filtering:
   • Compares with previous text using similarity ratio
   • Filters out short or duplicate text
4. Translation: Sends text to DeepL API for translation
5. Display: Shows translated text as transparent overlay

Troubleshooting

Common Issues

1. Tesseract not found:

   • Ensure Tesseract is installed and in PATH
   • Check path in settings matches actual location

2. DeepL API errors:

   • Verify API key is correct
   • Check API quota and billing status
   • Ensure internet connection is stable

3. No text detected:

   • Adjust OCR settings (try different modes)
   • Ensure selected area contains clear, readable text
   • Check screen resolution and DPI settings

4. Application won't start:

   • Check Python version (3.9+ required)
   • Verify all dependencies are installed
   • Check for missing language files

Debug Mode

Run with debug output:

python sfe.py

Check console output for detailed error messages and OCR results.

License

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

Support

For issues, questions, or feature requests:

• Open an issue on GitHub
• Check the troubleshooting section above
• Review existing issues for similar problems