Shirochi-stack / Glossarion

📚 Glossarion **Glossarion** is a comprehensive AI-powered translation suite for Korean, Japanese, and Chinese light novels, web novels, and manga. Built to transform EPUB files, raw .txt files, and manga images into high-quality, context-aware English translations. It supports multiple AI providers including OpenAI, Google Gemini, Anthropic Claude, DeepSeek, Mistral, and more (including local LLM's using ollama), with a modern GUI that gives you total control over every step of the translation process. --- 🏷️ Badges ✨ Key Features 📖 Novel Translation Engine • **Multi-Provider AI Support** • OpenAI (GPT-4, o1-preview, o1-mini) • Google Gemini (Flash, Pro, experimental models) • Anthropic Claude (Opus, Sonnet, Haiku) • DeepSeek, Mistral, Cohere, and more • **Advanced Context Management** • Rolling history window for consistent translations • Chunk-based processing for long chapters • Contextual memory with configurable depth • Full history export/import • **Smart Retry System** • AI Hunter duplicate detection • Automatic retry for timeouts and errors • Intelligent truncation detection and recovery • Rate limit handling with exponential backoff 🎌 Manga Translation Engine • **Dual API System: OCR + AI Translation** • Google Cloud Vision API for text detection (OCR) • Your chosen AI provider (OpenAI/Gemini/Claude/etc.) for actual translation • Requires BOTH: Google Cloud credentials AND your AI API key • **Visual Context-Aware Translation** • AI sees the full manga page image for accurate context • Full page context mode for multi-bubble consistency • Character expressions inform translation tone • **Best results with advanced models like o3** • **Advanced Text Rendering** • Customizable fonts, colors, and backgrounds • Text shadows and outlines for readability • Smart inpainting to remove original text • Preserve original art option • **Batch Processing** • Process entire manga chapters automatically • Stop/resume functionality • Progress tracking and error recovery 🧠 AI Hunter • **Advanced Duplicate Detection** • Machine learning-based content similarity analysis • Semantic fingerprinting using TF-IDF • Structural pattern recognition • Configurable sensitivity thresholds • **Smart Filtering** • Length ratio checks • Key phrase verification • Character name consistency • Statistical outlier detection 📓 Glossary System v2.0 • **Flexible Extraction** • Custom field support • Configurable prompts • Multi-language support (KR/JP/CN) • Duplicate merging strategies • **Advanced Management** • Field-specific trimming controls • Import/export with validation • Automatic backup system • JSON and Markdown formats 🛡️ Quality Assurance Suite • **Comprehensive Scanning** • Duplicate content detection • Non-English fragment identification • Spacing and formatting issues • Repetitive sentence patterns • **Multiple Report Formats** • Interactive HTML reports • Detailed JSON analysis • CSV exports for spreadsheets • Summary statistics 🖼️ Image Translation • **Smart Processing** • Auto-detection of text in images • Tall image splitting for reliable OCR • Batch processing with progress tracking • Context preservation across chunks 📚 File Format Support • **EPUB Processing** • Structure-preserving translation • Metadata and cover retention • Image gallery support • Clean HTML generation • **Text File Support** • Chapter detection algorithms • Custom delimiters • Encoding auto-detection • Format preservation 🖥️ Modern GUI Interface • **User-Friendly Design** • Dark/light theme support via ttkbootstrap • Real-time translation progress • Scrollable dialogs for all screens • Comprehensive logging system • **Advanced Configuration** • Per-language prompt profiles • Temperature and token controls • API endpoint customization • Batch size optimization --- 📦 Installation Prerequisites • Python 3.10 or higher • Windows 10/11 (for full feature support) Quick Start • Clone the repository: • Install dependencies: • Launch the GUI: Building Executable (Optional) --- 📋 Requirements --- 🔑 API Configuration Supported AI Providers Here's the updated API provider table based on the unified API client support: | Provider | Model Format | Example Models | |----------|--------------|----------------| | **OpenAI** | , , , | gpt-4, gpt-4-turbo, gpt-4o, o1-preview, o1-mini, o3, o4-mini | | **Google Gemini** | | gemini-1.5-flash, gemini-1.5-pro, gemini-2.0-flash, gemini-2.5-flash, gemini-2.5-pro | | **Anthropic** | | claude-3-opus, claude-3-sonnet, claude-3-haiku, claude-3.5-sonnet | | **DeepSeek** | | deepseek-chat, deepseek-coder | | **Mistral** | , | mistral-large, mistral-medium, open-mistral-7b | | **Cohere** | | command-r, command-r-plus | | **ElectronHub** | , | eh/gpt-4, eh/claude-3-opus, eh/gemini-2.5-flash, eh/yi-large | | **VertexAI** | , | vertex/gemini-2.5-flash, vertex/gemini-2.5-pro | **Groq** | , , | groq/llama-3.1-70b, groq/mixtral-8x7b | | **POE** | poe/* | poe/gemini-2.5-flash, poe/claude-3.5-instant, poe/anthropic-instant-v1 | | **Together AI** | , | together/llama-3-70b, meta-llama/Llama-3-70b | | **Fireworks** | , | fireworks/llama-v3-70b, accounts/fireworks/models/mixtral-8x7b | | OpenRouter | , | or/google/gemini-2.5-flash, or/openai/chatgpt-4o-latest | Chutes | | chutes/deepseek-ai/DeepSeek-V3.1 | **Perplexity** | , , | perplexity/llama-3.1-70b, pplx-7b-online | | **Anyscale** | , | anyscale/llama-3-70b, meta-llama/Llama-3-70b | | **Hugging Face** | | huggingface/meta-llama/Llama-3-70b | | **Replicate** | , | replicate/meta/llama-3-70b | | **AI21** | , , | ai21/jamba-1.5-large, j2-ultra | | **Voyage AI** | | voyage/voyage-3 | | **Reka** | | reka-flash, reka-core, reka-edge | | **xAI** | , | xai/grok-beta, grok-2 | | **LeptonAI** | , , | lepton/llama-3.1-70b, lepton/mixtral-8x7b | | **DeepInfra** | | deepinfra/meta-llama/Llama-3-70b | | **Qwen** | , | qwen-max, qwen-plus, qwen-turbo | | **Yi** | , , | yi-large, yi-medium, zero-one-ai/yi-34b | | **Moonshot** | | moonshot-v1-12…