Word Counter

Count exactly what you need

Type or paste text and every number updates live. The counter handles both space-separated languages and CJK text, where "words" are not delimited by spaces at all — so you get a character count you can trust for Chinese, Japanese and Korean copy as well.

What is counted

• Words — space-separated tokens (`Hello world` = 2 words).
• Characters — Unicode code points, so emoji and non-Latin letters count as one each.
• Characters without spaces — the number most forms actually ask for.
• Sentences — split on `.`, `!`, `?`, `。`, `！`, `？` and `…`.
• Paragraphs / lines — blank-line separated blocks and raw newlines.
• CJK characters — Han, Kana and Hangul, counted separately.
• Size (UTF-8) — real byte length, which is what a database or API limit means.
• Reading / speaking time — 225 wpm reading and 140 wpm speaking for Latin text, 400 and 220 characters per minute for CJK.

Where it helps

• Meeting a strict limit: tweets at 280 characters, meta descriptions around 155, ad headlines at 30/90, Chinese 摘要 often capped by characters.
• Writing abstracts where the journal counts characters, not words.
• Estimating video or podcast length from a script.
• Checking translation ratio between languages.

Privacy

Text is never uploaded — counting happens in the page, and closing the tab erases everything.

FAQ

How are words counted for Chinese text?

Chinese has no spaces, so token-based word counting is meaningless. The tool reports CJK characters separately and uses characters-per-minute for reading time.

Why do characters and bytes differ?

Characters count code points; bytes count UTF-8 storage. A Chinese character is 3 bytes in UTF-8, so the byte figure is much larger.

Which number do character limits use?

Most social platforms count characters (Twitter/X counts code points, where emoji may count as two). Database columns usually limit bytes.