关于此工具
Most word counters count Markdown symbols, LaTeX commands and code as words. This counter has four modes: plain text; Markdown, which strips headings, links, emphasis and code blocks; LaTeX, which removes commands, citations and math (counted separately) while keeping the text of \section{} or \textbf{}; and source code, which counts only comments and docstrings plus code, comment and blank lines. Words are segmented with Unicode rules, so Chinese, Japanese and Thai are counted correctly. Readability uses the right formula per language: Flesch for English, Fernández-Huerta for Spanish, Kandel-Moles for French, Amstad for German, Gulpease for Italian, Oborneva for Russian, Ateşman for Turkish and LIX for others.
与 LanguageHelper 上的所有工具一样,字数统计与可读性 完全在你的浏览器中运行。文本绝不会被上传、存储或用于训练,因此可放心处理机密文件、客户数据和私人消息。无需注册或安装,加载后即使离线也能继续使用。
使用方法
- Select the mode: plain text, Markdown, LaTeX or code.
- Paste or type — every number updates live.
- Adjust your reading speed for a personal reading-time estimate.
- Copy the statistics as plain text.
使用场景
- Check essay and assignment word limits accurately.
- Count words in Markdown blog posts without symbols and links.
- Measure LaTeX paper length without commands and equations.
- Estimate reading and speaking time for articles, talks and videos.
示例
处理前
# Title **Bold** [link](https://x.com)
处理后
3 words (symbols and URL excluded)
处理前
\section{Intro} Energy is $E=mc^2$ here.处理后
4 words, 1 math segment
使用技巧
Choose the mode that matches your source: Markdown for README files and blogs, LaTeX for academic papers, code for documentation coverage. Set the text language so readability uses the right formula — scores from different formulas are not directly comparable. Aim for shorter sentences and words to make text easier to read.
常见问题
How are words counted in Chinese and Japanese?
Chinese and Japanese don’t separate words with spaces, so the main count shows characters (the standard measure for these languages), and reading time uses characters per minute.
Which readability score is used?
The formula is chosen by language: Flesch Reading Ease and grade level (English), Fernández-Huerta (Spanish), Kandel & Moles (French), Amstad (German), Gulpease (Italian), Flesch-Douma (Dutch), Oborneva (Russian/Ukrainian), Ateşman (Turkish), FOG-PL (Polish), Nguyen–Henkin (Vietnamese), LIX (Indonesian, Romanian), characters per sentence (Chinese, Japanese — plus kanji ratio) and average sentence length (Arabic, Hindi, Korean, Thai), where no syllable-based formula is established.
What counts as a byte?
The size of your text encoded as UTF-8, the encoding used on the web and by most databases. Accented letters take 2 bytes, most CJK characters 3, and emoji 4.
How is reading time calculated?
Words divided by your chosen speed (230 words per minute by default). Speaking time assumes about 140 words per minute.
Is my document uploaded?
No. All counting happens in your browser as you type.
Are hyphenated words one word?
Words are segmented by Unicode rules, which treat hyphenated compounds such as “well-known” as separate parts in most browsers. Counts may therefore differ slightly from word processors.