• Real-time Translation with multiple OCR options (OneOCR, Windows OCR, PaddleOCR, EasyOCR, RapidOCR)
• AI-Powered Translation with Gemini, Groq, ChatGPT, Google Translate, Ollama, Mistral, LM Studio
• Smart Recognition with game context awareness and character name detection
• Flexible Display options with overlay and chat window
• Text-to-Speech feature with 4 backends: ElevenLabs (cloud), Google Cloud TTS (cloud), Windows TTS (local), and Supertonic (free, on-device, 31 languages, OpenRAIL-M model)
• Speech-to-Text functionality (Recognize speech from game audio and translate it)

RST supports four TTS backends. Pick the one that fits your needs:

Supertonic is a lightning-fast, on-device multilingual TTS by Supertone. It runs entirely on your CPU via ONNX Runtime - no cloud, no API key, no privacy concerns. After the one-time ~400 MB model download, it works completely offline.

How to enable:

1. Open Settings → TTS
2. Set TTS Service = Supertonic
3. Click Download model (≈400 MB, downloaded from Hugging Face)
4. Pick a voice style (M1, F1, M2, F2, M3, M3, M4, M5, F3, F4, F5)
5. Optionally tune Quality (denoise steps, default 8) and Speech speed (default 1.05)

Supported languages (31): en, ko, ja, ar, bg, cs, da, de, el, es, et, fi, fr, hi, hr, hu, id, it, lt, lv, nl, pl, pt, ro, ru, sk, sl, sv, tr, uk, vi plus na (language-agnostic auto-detect).

Notes:

• First SpeakText call after app launch may take 1-2s while the model loads; the app will warm the engine in the background if Supertonic is already selected, so subsequent calls are instant.
• License: SDK is MIT, model is OpenRAIL-M (free for commercial use with responsible-AI conditions).
• Model files: https://huggingface.co/Supertone/supertonic-3

• Windows 10+ and game in windowed/borderless mode
• NVIDIA GPU recommended but optional

1. Download from Releases and extract

1. Run rst.exe
2. Go to Settings → OCR: Select OneOCR
3. Go to Settings → Language: Choose languages
4. Go to Settings → Translation: Select Google Translate
5. Press button Select Window: Choose window which you want to capture
6. Press Alt+Q to select area, then Alt+F to turn on Overlay
7. Press Alt+G to start/stop

1. OCR Options: (Setup is only needed the first time the new OCR is chosen, no need to reinstall.)

   • Built-in: OneOCR, Windows OCR (no setup needed)
   • External: Click SetupOCR for PaddleOCR, RapidOCR, EasyOCR (5-15 min wait)

2. Translation Services:

   • No API needed: Google Translate
   • API required: Gemini, Groq, Mistral, ChatGPT (add keys in Settings)
   • Local options: Ollama, LM Studio

3. Start translating:

   • Click StartOCR (if using external OCR) and wait until it starts successfully (You will see a red notification line at the bottom right corner)
   • Press button Select Window: Choose window which you want to capture
   • Select area (Alt+Q) then turn on overlay (Alt+F)
   • Start translate (Alt+G)

RSTGameTranslation will automatically check for updates when you start it. If there's a new version, you'll see a notification asking if you want to download it. To update:

1. Download the latest version from the notification or from Releases
2. Close RSTGameTranslation if it's running
3. Extract the new files to your current installation folder
4. Done! Your settings and options will be preserved

• OCR: OneOCR or Windows OCR (built-in, no setup)
• Translation: Google Translate (no API key needed)

• OCR: PaddleOCR (Asian) or RapidOCR (Western) or EasyOCR
• Translation: Gemini Flash lite 2.5 (Need API key)
• Hardware: NVIDIA GPU recommended

• OCR: OneOCR or Windows OCR
• Translation: Ollama or LM Studio (100% local)

• Smaller areas = faster processing
• Add multiple API keys for failover
• First language download takes 1-2 minutes (external OCR)

Join our Discord for support and updates!

If you find RST useful and would like to support development, you can buy the author a coffee — thank you! 👋

This project stands on the shoulders of giants. We gratefully acknowledge the following open-source projects:

• WPF (Windows Presentation Foundation) - UI Framework

• PaddleOCR - Awesome multilingual OCR toolkits
• EasyOCR - Ready-to-use OCR with 80+ supported languages
• RapidOCR - Cross platform OCR library based on OnnxRuntime

• System.Text.Json - High-performance JSON serialization
• Hugging Face - For various AI models and datasets

• NAudio - Audio and MIDI library for .NET
• System.Speech - .NET Speech Synthesis Library
• Whisper.Net - .NET Speech To Text Library

• Python - Backend scripting
• PyTorch - Machine learning framework

BSD-style attribution - see LICENSE.md

Acknowledgments: Includes software developed by Seth A. Robinson - UGTLive

	Free code signing provided by SignPath.io, certificate by SignPath Foundation.