What this tool does
Text statistics breaks a piece of writing into the metrics people check most: characters, Chinese characters, English words, lines, sentences and paragraphs. Whether you write an article, revise a paper, translate, or draft short-video copy, you see the length at a glance.
Mixed Chinese/English text is counted correctly — Chinese by character, English by word — so a sandwiched English phrase will not throw the count off. Everything runs locally in your browser; nothing is uploaded.
How to use
- Paste or drop text Paste text into the box, or pick a .txt / .md / .csv / .json / .log file and drop it in (up to 5MB each).
- Toggle space counting if needed By default the character count includes spaces. Turn off “Count spaces as chars” to see visible characters only; the result updates live.
- Copy or download The result is plain text, one metric per line. Hit Copy to paste it, or Download to save a text-stats-*.txt file.
FAQ
How is mixed Chinese/English counted?
Chinese is counted by character and English by word; “characters” is the total code-point count (with or without spaces per the toggle). The three figures do not interfere.
How many characters is an emoji?
Counted as 1 by Unicode code point. Common emoji occupy a single code point, so they are not counted as two.
Does “characters” include spaces?
By default yes. Turning the toggle off switches the headline to the no-spaces value; both numbers appear in the report for reference.
Can it count Word / PDF?
This tool only takes plain-text files (.txt/.md/.csv/.json/.log). Export Word/PDF to plain text first, then paste it in.
Is my text uploaded?
No. Counting happens entirely on your device; the text never leaves the browser.
Can it count several files at once?
One file at a time for now. Batch aggregation is a later enhancement needing a dedicated output design.
Limits
Up to 5MB per file; .txt / .md / .csv / .json / .log. Sentence counting is heuristic (English abbreviations like “Mr.” may be miscounted as a sentence end) — fine for length checks, not for rigorous linguistics. No word frequency, readability or sentiment analysis.
Only plain-text files are handled; rich text (Word / PDF / formatted web) must be converted to plain text first, otherwise formatting marks are counted as characters.