Files
aistory-generate/README.md
T

5.7 KiB

TiniX Story 1.0

TiniX Story 1.0 is a production-ready, highly sophisticated intelligent novel generation system. It provides a comprehensive suite of AI-powered tools designed to assist authors, writers, and creators in brainstorming, outlining, drafting, rewriting, and polishing long-form fiction.

📖 Documentation / Tài liệu

Language Link
🇬🇧 English (Default) See below for English instructions
🇻🇳 Tiếng Việt (Vietnamese) locales/VI/huong_dan_su_dung.md

🌟 Core Features

  • ✍️ Smart Creation: Write novels from scratch. Generate creative titles, build detailed character profiles, design world settings, and construct comprehensive outlines automatically before generating chapters.
  • ⏸️ Resume Capability & Auto-Save: Fully supports pausing generation. Every chapter is saved automatically. You can resume writing from exactly where you left off via Project Management.
  • 🔄 Smart Rewrite: Upload your existing text and rewrite it using 17 high-quality, preset styles (e.g., Eastern Fantasy, Hard Sci-Fi, Urban Romance, Horror, etc.).
  • 📖 Smart Continue: Upload a partially written novel. The AI will analyze the context, characters, and world-building, and automatically continue the story maintaining your style and tone.
  • Novel Polish: Advanced polishing tools with 8 specific modes: General Polish, Find Errors, Improvement Suggestions, Direct Edit, Remove AI Tone, Enhance Details, Optimize Dialogue, and Improve Pacing.
  • 🎭 Diverse Web Novel Styles: Natively supports 30+ highly-specialized web novel genres and styles like Eastern Fantasy (Xianxia/Wuxia), System Face-slapping, Sci-Fi Mecha, Infinite Flow, Sweet Pet Romance, Slice of Life, eSports, Dark Fantasy, and more. AI will meticulously adapt vocabulary, pacing, and tone to fit the chosen sub-culture.
  • 🧠 Self-Reflection Check: A built-in critique loop where the AI generates a draft, reviews it against the outline to detect logic flaws or "AI tone", and rewrites a perfect version before saving.
  • 📄 Flexible File Parsing: Upload your own files (.txt, .pdf, .epub, .md, .docx). The system supports multiple parsing mechanisms including auto-split, split by word count, and split by custom Regex patterns.
  • 💾 Multi-Format Export: Export your completed or in-progress projects into Word (.docx), Plain Text (.txt), Markdown (.md), or HTML (.html) with a single click.
  • 📂 Project Management: Manage multiple novel projects simultaneously. Track completion percentage, chapter counts, update dates, and manage generation cache.
  • ⚙️ Flexible Configuration (Multi-API): Add multiple API backends (OpenAI compatible, Anthropic, Google, Ollama). Set default providers, test connections, and configure individual generation parameters (Temperature, Top P, Max Tokens) per backend.
  • 🛡️ Built-in Resilience:
    • Error Retry: Automatically retries if a generated chapter fails or returns 0 words.
    • Port Auto-detection: Automatically finds available ports if the default Gradio port (7860) is occupied.
    • Rate Limiting Handling: Built-in token bucket algorithms to gracefully handle API rate limits.

🔧 Technology Stack & Architecture

  • Backend: Python 3.11+, leveraging async operations and thread-safe concurrency.
  • Frontend: Gradio 4.0+ for a responsive, component-driven Web UI.
  • Persistence: SQLite for robust project metadata storage, combined with standard file I/O for chapter content and caching.
  • Caching: Intelligent caching system stores generated chapters and context summaries to drastically reduce redundant API calls and save costs.
  • Internationalization (i18n): Fully dynamic string loading supporting multiple locales dynamically switchable via environment variables.

🚀 Quick Start Guide

Prerequisites

  • Python 3.8 or higher installed on your system.
  • Git (optional, for cloning the repository).

Installation & Setup

  1. Clone or download the repository:

    git clone https://github.com/tinix-ai/tinix-story.git
    cd tinix-story
    
  2. Create a virtual environment (Recommended):

    python -m venv venv
    # On Windows
    venv\Scripts\activate
    # On macOS/Linux
    source venv/bin/activate
    
  3. Install Dependencies:

    pip install -r requirements.txt
    

    (Optional) To enable PDF or EPUB parsing:

    pip install PyMuPDF ebooklib beautifulsoup4
    
  4. Launch the Application:

    python app.py
    

    The terminal will display the local URL (usually http://127.0.0.1:7860). Open this link in your browser.


🌐 i18n (Internationalization)

The application supports multiple languages out of the box. Language files are stored in the locales/ directory:

locales/
├── i18n.py               # i18n helper module
├── __init__.py           
├── EN/                   # English (Default fallback)
│   └── messages.json     
└── VI/                   # Vietnamese
    ├── messages.json     
    └── huong_dan_su_dung.md         

Switching Language

Set the APP_LANGUAGE environment variable before starting the app.

Windows (Command Prompt):

set APP_LANGUAGE=VI
python app.py

Windows (PowerShell):

$env:APP_LANGUAGE="VI"
python app.py

macOS/Linux:

APP_LANGUAGE=VI python app.py

💻 Contribution & Support

We welcome pull requests and issue reports! If you encounter connection issues, ensure your API keys are valid and double check the logs located in the logs/ directory.

License

This project is licensed under the MIT License.