Dota 2 Chat Translator

This application captures a specified region of your screen, performs Optical Character Recognition (OCR) on it, and translates any recognized text into English. It is designed to translate in-game chat from Dota 2 in real-time.

This version uses a hybrid approach:

• Local OCR: Text recognition is handled locally using the Tesseract OCR engine. This is fast, free, and has no usage limits.
• Cloud Translation: The recognized text is translated using the Google Cloud Translation API to ensure high-quality translations.

Installation

There are three main steps to get the application running: installing Tesseract, setting up the Python environment, and configuring Google Cloud.

1. Install Tesseract OCR

This application depends on Tesseract OCR engine. You must install it on your system.

• Platform: Windows
• Installer: You can find the official Tesseract installers from the University of Mannheim.
  • Link: https://github.com/UB-Mannheim/tesseract/wiki
• Instructions:
  1. Download and run the installer for your system (e.g., tesseract-ocr-w64-setup-v5.x.x.exe).
  2. Important: During installation, make sure to select the language packs for the languages you expect to encounter. At a minimum, for Dota 2 it is recommended to install:
     • English (eng)
     • Russian (rus)
     • Spanish (spa)
     • Portuguese (por)
     • Simplified Chinese (chi_sim)
  3. Note the installation path. The default is usually C:\Program Files\Tesseract-OCR. If you install it here, the application should find it automatically. If you choose a different path, you may need to edit ocr_service.py and set the path manually at the top of the file.

2. Set Up Python Environment

Clone this repository and install the required Python packages.

# Clone the repository
git clone <repository-url>
cd Dota2ChatTranslater

# It is recommended to use a virtual environment
python -m venv venv
venv\Scripts\activate

# Install dependencies
pip install -r requirements.txt

3. Configure Google Cloud for Translation

Since translation is handled by Google Cloud, you still need a Google Cloud Project to use the Translation API.

1. Create a Google Cloud Project: Follow the instructions on the Google Cloud Console.
2. Enable the "Cloud Translation API": In your project, go to the API Library and enable it.
3. Create Credentials:
   • Go to "Credentials" in the "APIs & Services" section.
   • Create a new credential of type "OAuth 2.0 Client ID".
   • Select "Desktop app" as the application type.
   • Download the JSON file containing your credentials.
4. Place Credentials File: Rename the downloaded file to client_secret.json and place it in the root directory of this project.
5. Set Project ID:
   • Run the application.
   • Go to Settings -> Google Cloud.
   • Paste your Google Cloud Project ID into the entry box and click "Save Project ID".

SECURITY WARNING: The client_secret.json file contains credentials for your Google Cloud Project. NEVER commit this file to a public version control repository (like GitHub). Always ensure it's listed in your .gitignore to prevent accidental exposure.

How to Run

Execute the run.bat script or run the main python file:

python main.py

Usage

1. Authorize Google Cloud: On first run (or after saving a new Project ID), click the "Authorize" button in the Settings menu. This will open a browser window asking you to log in with your Google account and grant permission.
2. Select Chat Region:
   • In the app's settings, click "Select Region".
   • Your screen will dim. Click and drag to draw a box around the area where chat messages appear in Dota 2.
3. Translate:
   • Once the region is set and Google Cloud is authorized, you can press your configured hotkey (default is <f8>) to capture the chat region.
   • The application will process the image, and any translated text will appear in the main window.
4. Settings: You can change the hotkey, theme, and text font size from the Settings menu.