Word, Character and Sentence Counter

How to use it

1. 1Paste your text into the box marked "Your text"Paste or type anything — an article, an essay, a description you are fitting into a limit. Everything happens inside your browser — nothing is uploaded, so there is nothing for us to store, log or lose.
2. 2Read the result as it updatesThere is no button to press. Every count updates on each keystroke. The output recomputes every time you change the text or one of the options above it.
3. 3Read the counts that differ from each otherCharacters with and without spaces are both shown, because the platform imposing your limit will have picked one and rarely says which. Words, sentences and paragraphs are counted separately.
4. 4Adjust the reading speed if you need a timeReading time is words divided by a rate you can set. The default suits general prose; technical writing is genuinely slower, and a figure from the wrong rate is worse than none.
5. 5Expect emoji and non-Latin scripts to be counted properlyCharacters are counted as people see them rather than as storage units, so an emoji is one character here even though it occupies several bytes.
6. 6Copy or download the resultThe buttons under the output put it on your clipboard or save it as a file, and each one says exactly what it will copy or download. Any note about something the tool changed or deliberately left alone is shown with the result rather than hidden.

About this tool

Counts words, characters with and without spaces, sentences, paragraphs, lines and reading time, updating as you type. Nothing is uploaded, so it is safe for a draft you have not published or a document you should not paste into a random website.

Counting looks trivial and is not. Emoji, accented letters and scripts written without spaces all break the obvious implementation, and they break it silently — the number looks reasonable and is wrong.

Understanding the result

Two character counts are shown because "characters" means two different things. A reader counts glyphs; a database column and most APIs count UTF-16 code units. An emoji is one glyph and two code units, so a caption that "fits" can still be rejected.

Words are found with the Unicode word-breaking algorithm rather than by splitting on spaces, so Chinese, Japanese and Thai — which are written without spaces — are counted properly instead of coming back as one word.

Sentence detection knows that "3.50" is not the end of a sentence, and a short abbreviation list keeps "Dr. Smith arrived." as one. Reading time uses 238 words per minute for silent reading, which you can change.

Example

The quick brown fox jumps over the lazy dog.

9 words · 44 characters · 36 without spaces · 1 sentence

Limitations

• Sentence counting is a heuristic. An unusual abbreviation, a list without full stops, or dialogue with heavy punctuation can shift the count by one or two.
• Reading speed varies enormously between readers and between material. Treat the estimate as a rough guide, not a measurement.
• Hyphenated compounds are counted as one word, matching what a word processor reports. Tools that split on hyphens will give a higher number for the same text.
• For a URL, a token or code, the word and sentence counts are meaningless. The tool says so when it detects that, and the character and byte counts remain correct.

Questions