Text Tools

Word Counter

A word counter reports how many words, sentences and paragraphs a piece of writing contains, and estimates how long it takes to read aloud or in silence. This one counts sentences using the Unicode text rules, so an abbreviation like “Dr.” or a price like 9.99 does not get mistaken for the end of a sentence. Everything is measured in your browser as you type.

Runs entirely in your browser — your text is never uploaded.

Your text

Words: 0
Characters: 0
No spaces: 0
Sentences: 0
Paragraphs: 0
Lines: 0

Reading time

—

Silent reading, 238 words a minute

Speaking time

—

Read aloud, 130 words a minute

How to

How to use the Word Counter

1
Add your text
Type directly into the box, paste from the clipboard, or drop in a .txt file. Counting starts immediately — there is no button to press.
2
Read the counts
Words, characters, sentences and paragraphs update on every keystroke, alongside estimated reading and speaking times.
3
Check the sentence length
Average words per sentence and the longest sentence tell you whether the writing will feel heavy. Anything past roughly 25 words tends to lose readers.
4
Look at your repeated words
The most-used list filters out “the”, “and” and similar, leaving the words you actually lean on. Copy the summary when you are done.

Examples

Word Counter examples

An abbreviation is not a sentence end

Input: Dr. Smith arrived. He paid 9.99 for it.
Output: 2 sentences, 8 words

Splitting on every full stop would report four sentences here — one for “Dr.”, one for the decimal in 9.99. Sentence detection skips abbreviations and numbers.

A blog post

Input: 1,400 words of draft copy
Output: ≈ 6 min read, ≈ 11 min spoken

Reading time uses 238 words per minute, the average for adults reading prose silently. Speaking time uses 130, which is a comfortable presentation pace.

Text without spaces

Input: 你好世界你好
Output: 3 words

Chinese and Japanese do not separate words with spaces. Splitting on whitespace would call this one word; the Unicode word rules segment it properly.

Why use it

What the Word Counter gives you

Counts that hold up

Sentence detection ignores abbreviations, decimals, initials and URLs, so the number matches what you would get counting by hand.

Reading time you can quote

Estimates come from measured adult reading and speaking rates rather than a round number, and both are shown so you can pick the relevant one.

Repetition made visible

The most-used words list strips out filler, which makes an overused phrase obvious in a way that reading the draft again rarely does.

Nothing is uploaded

Counting happens in the page. An unpublished manuscript or a confidential report never leaves your machine.

Good to know

Word Counter limitations

Reading time is an average. Dense technical writing reads slower and light fiction faster, sometimes by half.
Sentence detection uses a list of common abbreviations. An unusual one followed by a capital letter may still be read as a sentence break.
Only plain text is read. Pasting from a word processor keeps the words but drops formatting, footnotes and comments.
Very large documents past a few hundred thousand words will make typing feel sluggish, since every keystroke recounts.

Summary

Word Counter in short

Words, sentences, paragraphs, reading time and speaking time update as you type.
Sentence counting skips abbreviations, decimals and initials instead of splitting on every full stop.
Average and longest sentence length flag writing that has become hard to follow.
Chinese, Japanese and other scripts without spaces are segmented correctly.
Everything runs in your browser, so nothing is uploaded.

FAQ

Word Counter questions

How is a word counted?

By the Unicode word rules, which treat a hyphenated compound like “well-known” as one word and a contraction like “don’t” as one word, while excluding stray punctuation. It matches what a word processor reports for ordinary prose.

Why does my sentence count differ from other tools?

Most tools split on every full stop, so “Dr. Smith paid 9.99” counts as three sentences. Abbreviations, decimals, initials and web addresses are skipped here, which usually gives a lower and more accurate number.

Where does the reading time come from?

238 words per minute, the mean for adults reading English prose silently, from a meta-analysis of reading rate studies. Speaking time uses 130 words per minute, roughly the pace of an unhurried presentation.

What counts as a paragraph?

A block of text separated from the next by a blank line. A single line break inside a block does not start a new paragraph, which matches how word processors and Markdown both behave.

Why are common words missing from the most-used list?

Words like “the”, “and” and “of” top the list in every text ever written, so they are filtered out. What remains is the vocabulary specific to your writing, which is where repetition actually shows.

Does it work with languages other than English?

Yes. Accented words are counted and listed intact, and scripts without spaces between words — Chinese, Japanese, Thai — are segmented by the Unicode rules rather than being treated as one long word.

What is a good average sentence length?

Between 15 and 20 words suits most general writing. Past about 25 the reader starts having to hold too much in mind at once, which is why the longest sentence is shown separately — one runaway sentence can undo an otherwise readable page.

Can I load a file instead of pasting?

Yes, plain .txt files can be dropped straight in. Word documents and PDFs are not read, because their text is wrapped in formatting this tool deliberately does not parse.

Is my writing sent anywhere?

No. The text stays in the page and is never transmitted, stored or logged. Closing the tab discards it, which is what you want for an unpublished draft or anything under embargo.

Discover

Related text tools

Character CounterCount characters against Twitter, Instagram, SMS and meta tag limits. Counts emoji as one character, the way readers see them.Case ConverterConvert text between UPPERCASE, Title Case, camelCase, snake_case and kebab-case. Splits existing camelCase properly and keeps accents.