Skip to content

README.md

Latest commit

 

History

History
215 lines (158 loc) · 5.27 KB

README.md

File metadata and controls

215 lines (158 loc) · 5.27 KB

Text Humanizer 🤖➡️👤

Stop sounding like a bot. Instantly check your text's "Humanity Score".

Python Version License Code style: black

English | 中文

English

Features

  • 📊 Instant Readability Score - Get a 0-100 score showing how "human" your text sounds
  • 🎨 Beautiful Terminal UI - Color-coded gauge from 🔴 Robot to 🟢 Human
  • 📝 Multiple Formats - Supports Markdown (.md) and plain text (.txt) files
  • 🚨 AI Word Detection - Highlights common AI-generated words like "delve", "testament", "tapestry"
  • ⚡ Fast & Lightweight - No API calls, runs completely offline

Quick Start

Installation

pip install text-humanizer

Or install from source:

git clone https://github.com/PerryLink/text-humanizer.git
cd text-humanizer
poetry install

Usage

Check a file:

check-human ./draft.md

Pipe from clipboard (macOS):

pbpaste | check-human -

Pipe from clipboard (Windows):

powershell Get-Clipboard | check-human -

How It Works

The tool uses proven readability algorithms:

  1. Flesch Reading Ease Score - Higher scores = easier to read

    • 0-30: Very difficult (Academic/Robot)
    • 30-60: Difficult (Standard)
    • 60-100: Easy (Conversational/Human)

  2. Average Sentence Length - Shorter sentences = more natural

    • Penalty applied if average > 25 words

  3. AI Word Detection - Flags common AI-generated words that make text sound robotic

Project Structure

text-humanizer/
├── src/
│   └── text_humanizer/
│       ├── __init__.py
│       ├── __main__.py
│       ├── analyzer.py      # Core analysis logic
│       ├── cleaner.py       # Text preprocessing
│       ├── cli.py           # Command-line interface
│       └── ui.py            # Terminal UI rendering
├── tests/
│   ├── test_analyzer.py
│   └── test_cleaner.py
├── examples/
│   ├── sample_human.txt
│   ├── sample_robot.txt
│   └── sample_markdown.md
├── pyproject.toml
├── README.md
├── LICENSE
└── CONTRIBUTING.md

Tech Stack

  • Python 3.8+ - Core language
  • textstat - Readability metrics calculation
  • rich - Beautiful terminal output
  • typer - CLI framework
  • pytest - Testing framework
  • black - Code formatting

License

This project is licensed under the Apache License 2.0 - see the LICENSE file for details.

Copyright 2026 Chance Dean (novelnexusai@outlook.com)

Contributing

Contributions are welcome! Please see CONTRIBUTING.md for details.

中文

功能特性

  • 📊 即时可读性评分 - 获得 0-100 分的"人性化"评分
  • 🎨 精美终端界面 - 从 🔴 机器人到 🟢 人类的彩色评分条
  • 📝 多格式支持 - 支持 Markdown (.md) 和纯文本 (.txt) 文件
  • 🚨 AI 词汇检测 - 高亮显示常见的 AI 生成词汇,如 "delve"、"testament"、"tapestry"
  • ⚡ 快速轻量 - 无需 API 调用,完全离线运行

快速开始

安装

pip install text-humanizer

或从源码安装:

git clone https://github.com/PerryLink/text-humanizer.git
cd text-humanizer
poetry install

使用方法

检查文件:

check-human ./draft.md

从剪贴板读取(macOS):

pbpaste | check-human -

从剪贴板读取(Windows):

powershell Get-Clipboard | check-human -

工作原理

本工具使用经过验证的可读性算法:

  1. Flesch 阅读易度评分 - 分数越高 = 越易读

    • 0-30:非常困难(学术/机器人)
    • 30-60:困难(标准)
    • 60-100:简单(对话/人类)

  2. 平均句子长度 - 句子越短 = 越自然

    • 平均超过 25 个单词会被扣分

  3. AI 词汇检测 - 标记使文本听起来像机器人的常见 AI 生成词汇

项目结构

text-humanizer/
├── src/
│   └── text_humanizer/
│       ├── __init__.py
│       ├── __main__.py
│       ├── analyzer.py      # 核心分析逻辑
│       ├── cleaner.py       # 文本预处理
│       ├── cli.py           # 命令行界面
│       └── ui.py            # 终端 UI 渲染
├── tests/
│   ├── test_analyzer.py
│   └── test_cleaner.py
├── examples/
│   ├── sample_human.txt
│   ├── sample_robot.txt
│   └── sample_markdown.md
├── pyproject.toml
├── README.md
├── LICENSE
└── CONTRIBUTING.md

技术栈

  • Python 3.8+ - 核心语言
  • textstat - 可读性指标计算
  • rich - 精美终端输出
  • typer - CLI 框架
  • pytest - 测试框架
  • black - 代码格式化

许可证

本项目采用 Apache License 2.0 许可证 - 详见 LICENSE 文件。

版权所有 2026 Chance Dean (novelnexusai@outlook.com)

贡献

欢迎贡献!详情请参阅 CONTRIBUTING.md。