{"id":49603105,"url":"https://github.com/tomkam1702/ocr-translator","last_synced_at":"2026-05-09T11:14:20.505Z","repository":{"id":300277169,"uuid":"977525604","full_name":"tomkam1702/OCR-Translator","owner":"tomkam1702","description":"🎮 Real-time game subtitle translator with AI-powered OCR. Context-aware translation for 20+ languages. Free offline models + dirt cheap APIs. Perfect for gaming in foreign languages!","archived":false,"fork":false,"pushed_at":"2026-05-02T15:06:29.000Z","size":30875,"stargazers_count":43,"open_issues_count":7,"forks_count":8,"subscribers_count":2,"default_branch":"main","last_synced_at":"2026-05-02T16:29:32.077Z","etag":null,"topics":["deepl","game-translation","games","gemini","google-translate","in-game","language-learning-game","llm","marianmt","ocr","python","real-time-translation","subtitle-translation","subtitles","tesseract","translation","translator"],"latest_commit_sha":null,"homepage":"https://tomkam1702.github.io/OCR-Translator/","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"other","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/tomkam1702.png","metadata":{"files":{"readme":"README.md","changelog":"CHANGELOG.md","contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2025-05-04T12:09:42.000Z","updated_at":"2026-05-02T15:06:32.000Z","dependencies_parsed_at":null,"dependency_job_id":"ef411963-8797-463c-aab0-5a5b96833e0d","html_url":"https://github.com/tomkam1702/OCR-Translator","commit_stats":null,"previous_names":["tomkam1702/ocr-translator"],"tags_count":20,"template":false,"template_full_name":null,"purl":"pkg:github/tomkam1702/OCR-Translator","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/tomkam1702%2FOCR-Translator","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/tomkam1702%2FOCR-Translator/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/tomkam1702%2FOCR-Translator/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/tomkam1702%2FOCR-Translator/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/tomkam1702","download_url":"https://codeload.github.com/tomkam1702/OCR-Translator/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/tomkam1702%2FOCR-Translator/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":32602730,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-03T22:12:39.696Z","status":"online","status_checked_at":"2026-05-04T02:00:06.625Z","response_time":58,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["deepl","game-translation","games","gemini","google-translate","in-game","language-learning-game","llm","marianmt","ocr","python","real-time-translation","subtitle-translation","subtitles","tesseract","translation","translator"],"created_at":"2026-05-04T10:02:59.113Z","updated_at":"2026-05-04T10:03:02.626Z","avatar_url":"https://github.com/tomkam1702.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Game-Changing Translator v4\nCopyright © 2025-2026 Tomasz Kamiński\u003cbr\u003e**Author of [![OHLC Forge Logo](assets/OHLC_Forge_Logo_small.png)](https://github.com/tomkam1702/OHLC-Forge) [OHLC Forge](https://github.com/tomkam1702/OHLC-Forge) – Professional tool for crypto traders**\n\n![Game-Changing Translator Logo](docs/screenshots/readme_screen.jpg)\n\n## 🌟 Overview\n\n**Game-Changing Translator** is a state-of-the-art desktop application designed for real-time screen translation. Using powerful AI-driven OCR and context-aware engines, it captures text from any part of your screen—be it a game, a movie, or a document—and translates it instantly into over 100 languages. \n\nWhether you're exploring the world of *The Witcher*, learning a new language through entertainment, or simply need to understand content that can't be copied, GCT v4 provides a seamless, immersive experience with floating overlays that stay on top of your content.\n\n---\n\n## 🎬 Featured Game Demonstrations\n\n\u003cdiv align=\"center\"\u003e\n  \u003ctable\u003e\n    \u003ctr\u003e\n      \u003ctd align=\"center\" width=\"33%\"\u003e\n        \u003ca href=\"https://youtu.be/0bMoL1pR7tM\"\u003e\n          \u003cimg src=\"https://img.youtube.com/vi/0bMoL1pR7tM/maxresdefault.jpg\" alt=\"The Witcher 3 Real-Time Translation Demo\" style=\"width: 100%; border-radius: 10px; box-shadow: 0 8px 24px rgba(0,0,0,0.3);\"\u003e\n        \u003c/a\u003e\n        \u003cbr\u003e\n        \u003cstrong\u003e🧙‍♂️ \u003ca href=\"https://youtu.be/0bMoL1pR7tM\"\u003eThe Witcher 3\u003c/a\u003e\u003c/strong\u003e\n        \u003cbr\u003e\n        \u003cem\u003eRevolutionary AI OCR \u0026 natural Polish-to-English translation\u003c/em\u003e\n      \u003c/td\u003e\n      \u003ctd align=\"center\" width=\"33%\"\u003e\n        \u003ca href=\"https://youtu.be/Iy4bIr06Ae4\"\u003e\n          \u003cimg src=\"https://img.youtube.com/vi/Iy4bIr06Ae4/maxresdefault.jpg\" alt=\"Kingdom Come: Deliverance II Real-Time Translation Demo\" style=\"width: 100%; border-radius: 10px; box-shadow: 0 8px 24px rgba(0,0,0,0.3);\"\u003e\n        \u003c/a\u003e\n        \u003cbr\u003e\n        \u003cstrong\u003e⚔️ \u003ca href=\"https://youtu.be/Iy4bIr06Ae4\"\u003eKingdom Come: Deliverance II\u003c/a\u003e\u003c/strong\u003e\n        \u003cbr\u003e\n        \u003cem\u003eCzech-to-English Translation\u003c/em\u003e\n      \u003c/td\u003e\n      \u003ctd align=\"center\" width=\"33%\"\u003e\n        \u003ca href=\"https://youtu.be/rCsfY6Zsmps\"\u003e\n          \u003cimg src=\"https://img.youtube.com/vi/rCsfY6Zsmps/maxresdefault.jpg\" alt=\"Star Wars: The Old Republic Real-Time Translation Demo\" style=\"width: 100%; border-radius: 10px; box-shadow: 0 8px 24px rgba(0,0,0,0.3);\"\u003e\n        \u003c/a\u003e\n        \u003cbr\u003e\n        \u003cstrong\u003e🌌 \u003ca href=\"https://youtu.be/rCsfY6Zsmps\"\u003eStar Wars: The Old Republic\u003c/a\u003e\u003c/strong\u003e\n        \u003cbr\u003e\n        \u003cem\u003eFrench-to-English Translation\u003c/em\u003e\n      \u003c/td\u003e\n    \u003c/tr\u003e\n  \u003c/table\u003e\n\u003c/div\u003e\n\n---\n\n## 🪄 Introducing Version 4: A New Era of Translation\n\n**Version 4** is not just an update — it is a complete architectural redesign. Built with a brand-new engine from the ground up, it offers a smoother, more responsive experience and a state-of-the-art interface tailored for modern gaming.\n\n### ✨ Redesigned GUI \u0026 Simple Mode\nThe new interface offers two distinct modes: **Simple** for hassle-free operation and **Custom** for granular control.\n\n\u003e [!TIP]\n\u003e **Update v4.0.1 - Mouse Click-Through**: Overlays now automatically become \"transparent\" to mouse clicks during translation. This allows you to interact with the game underneath without losing focus or accidentally dragging frames.\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"docs/user_manual/Simple.png\" width=\"80%\" alt=\"Simple Mode GUI\"\u003e\n\u003c/p\u003e\n\n### ⚡ Up and running in 3 simple steps:\n1.  **Set your target language** – Source language is auto-detected in Simple mode.\n2.  **Enter your API keys** – Mandatory Gemini key and optional DeepL key.\n3.  **Select areas and start** – Position your Source and Target overlays and hit **Start (~)**.\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"docs/user_manual/japanese01.png\" width=\"80%\" alt=\"Japanese Frame Setup 1\"\u003e\n  \u003cimg src=\"docs/user_manual/japanese02b.png\" width=\"80%\" alt=\"Japanese Frame Setup 2\"\u003e\n  \u003cimg src=\"docs/user_manual/japanese03b.png\" width=\"80%\" alt=\"Japanese Frame Setup 3\"\u003e\n\u003c/p\u003e\n\n---\n\n## 🛠️ Features\n\n### 🎁 FREE Features\n*   **Gemini AI OCR**: Industry-first AI-powered text recognition that handles stylised fonts and low-contrast backgrounds.\n*   **Gemini Translation**: Top-quality, context-aware translation in over 100 languages.\n*   **Sliding Context Window**: Remembers up to 5 previous subtitles to maintain narrative coherence.\n*   **Cost Monitoring**: Real-time token-level analytics and cost tracking.\n*   **Two-Tier Caching**: In-memory and file-based caching to save on API costs.\n*   **Native RTL Support**: Flawless bidirectional rendering for Arabic, Hebrew, Persian, etc.\n*   **Translation Prompt**: Inject custom instructions to define tone or game-specific context.\n*   **API Logs**: Comprehensive dual-layer logging for both OCR and Translation.\n\n### 👑 PRO Features\n*   **DeepL Translation**: Elite precision for Japanese, Chinese, and European scripts with free context subtitles.\n*   **Find Subtitles**: Automatically scans the screen to detect and lock onto subtitle areas.\n*   **Target on Source**: Automatically overlays the translation directly onto the original subtitle area.\n*   **Scan Wider**: Expands the capture area to prevent word truncation and AI hallucinations.\n*   **OCR Prompt**: Custom instructions for Gemini OCR to filter HUD elements or speaker names.\n*   **Custom Appearance**: Full control over background and text colours with native pickers.\n\n---\n\n## 💎 PRO Feature Showcase\n\n#### 🪟 Target on Source \u0026 🔍 Find Subtitles\nThese features work together to create an \"invisible\" translator. **Find Subtitles** scans your screen to detect where text appears, while **Target on Source** places the translation directly on top of the original subtitles.\n\n\u003cdiv align=\"center\"\u003e\n\n**Step 1: Initial Scan**\n\n\u003cimg src=\"docs/user_manual/GCT_screenshot_2026-04-20_22-50-52.png\" width=\"90%\" alt=\"Initial Scan\"\u003e\n\n*The capture frame starts at a default size.*\n\n**Step 2: Adaptive Growth**\n\n\u003cimg src=\"docs/user_manual/GCT_screenshot_2026-04-20_22-50-56.png\" width=\"90%\" alt=\"Adaptive Growth\"\u003e\n\n*The frame expands in real-time to fit longer subtitles.*\n\n**Step 3: Locked \u0026 Immersive**\n\n\u003cimg src=\"docs/user_manual/GCT_screenshot_2026-04-20_22-51-02.png\" width=\"90%\" alt=\"Locked \u0026 Immersive\"\u003e\n\n*The frame locks at the perfect width, providing a seamless translation overlay.*\n\n\u003c/div\u003e\n\n\n\n### ✍️ OCR Prompt\nTake direct control over what the AI \"sees\". Filter out complex HUD elements, minimaps, and speaker names to focus only on the dialogue.\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"docs/user_manual/OCR_custom_prompt_example.png\" width=\"80%\" alt=\"OCR Prompt Demo\"\u003e\n  \u003cbr\u003e\n  \u003cstrong\u003eResult: \"Bang. Brain all up the walls.\"\u003c/strong\u003e – the speaker name \"Vincenzo:\" is stripped, and all HUD elements are ignored. The debug log on the right confirms the exact OCR output.\n\u003c/p\u003e\n\n---\n\n## ⚠️ Important Information\n\n\u003e [!CAUTION]\n\u003e ### Licensing \u0026 API Costs – Important Note\n\u003e Please distinguish between the **GCT Software Licence** and **Third-Party API Costs**:\n\u003e *   ![](https://img.shields.io/badge/FREE-f0fdf4?style=flat-square\u0026logoColor=166534) **Features:** These are unlocked in the GCT software for everyone. However, using them requires a connection to the **Google Gemini API**, which carries its own usage costs.\n\u003e *   ![](https://img.shields.io/badge/PRO-fefce8?style=flat-square\u0026logoColor=92400e) **Features:** These require a one-time purchase of a **GCT PRO Licence** to unlock advanced functionality within the program. This fee covers only software access and does **not** include or cover any API costs.\n\u003e *   **Independent API Services:** GCT is a professional interface for AI services provided by **Google** and **DeepL**. These are independent commercial entities. You are responsible for all costs incurred through their respective APIs (with the exception of the 500,000 characters/month provided in DeepL's free tier).\n\u003e *   **No Affiliation:** Game-Changing Translator and its author are entirely independent and have **no affiliation** with Google or DeepL. GCT is a tool designed to facilitate the use of these third-party paid services.\n\n\u003e [!WARNING]\n\u003e ### Compatibility \u0026 Version 4 Requirements\n\u003e Please review these technical requirements before proceeding:\n\u003e *   **API-Only Architecture:** Version 4 is built entirely around third-party AI APIs. Operation is **impossible** without a valid **Gemini API Key**.\n\u003e *   **Stay on v3.9.6:** If you do not have (and do not plan to obtain) a Gemini API key, you should **not** update to version 4. Please remain on **version 3.9.6**, which is the final release supporting free offline OCR (Tesseract) and offline translation models (MarianMT).\n\u003e *   **Try Before You Buy:** Do not purchase the **GCT PRO Licence** before thoroughly testing the **FREE** version. Ensure the software works correctly on your system and that the Gemini-powered OCR and translation meet your expectations.\n\n\u003e [!NOTE]\n\u003e ### Updating from v3.9.6 or v4.0.0\n\u003e Due to significant architectural changes in v4, automatic updates from versions **3.9.6** and **4.0.0** are not supported. If you are currently on either of these versions, please download and install the latest release **manually** from the [Releases](https://github.com/tomkam1702/OCR-Translator/releases) page. Automatic updates will work normally starting from version **4.0.1**.\n\n---\n\n## 🛑 Deprecated Features\n\nVersion 4 marks a significant shift towards high-quality AI-driven workflows. As a result, the following features have been retired:\n*   **Tesseract** OCR and **MarianMT** translation: These traditional/local technologies no longer keep up with the state-of-the-art results provided by modern AI.\n*   **OpenAI** and **Google Translate** support: Retired to streamline the application towards the most effective and cost-efficient engines.\n\n---\n\n## 🔗 Links\n\n*   📖 **[Full User Manual](https://tomkam1702.github.io/OCR-Translator/docs/user-manual.html)**\n*   💰 **[Get GCT PRO on Gumroad](https://tomkam17.gumroad.com/l/gct)**\n*   🌐 **[Official Website](https://tomkam1702.github.io/OCR-Translator/)**\n\n---\n\n## 📜 Licence\n\nThis project is proprietary software. The source code is provided for educational and evaluation purposes under a restrictive **End User License Agreement (EULA)**.\n\n**You may:**\n*   Use the Free Edition for personal and evaluation purposes.\n*   Review the source code for learning.\n\n**You may NOT:**\n*   Redistribute or sell the software.\n*   Modify the software to bypass PRO license checks.\n*   Use the code for commercial gain without permission.\n\nFor full details, see the [LICENSE](LICENSE) file.\n\n[![License: EULA](https://img.shields.io/badge/License-Proprietary%20EULA-orange.svg)](LICENSE)\n[![Built with PySide6](https://img.shields.io/badge/Built%20with-PySide6%20(LGPL%20v3)-blue.svg)](https://www.qt.io/qt-for-python)\n\n---\n### 🔗 Author's Portfolio\n[\u003cimg src=\"https://raw.githubusercontent.com/tomkam1702/OHLC-Forge/main/images/OHLC_Forge_Logo.png\" width=\"60\" height=\"60\" align=\"absmiddle\"\u003e](https://github.com/tomkam1702/OHLC-Forge) **[OHLC Forge](https://github.com/tomkam1702/OHLC-Forge)** — Professional OHLC daily data reconstruction for Binance and Bybit.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Ftomkam1702%2Focr-translator","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Ftomkam1702%2Focr-translator","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Ftomkam1702%2Focr-translator/lists"}