Skip to main content

Text Analyzer

Analyze text with detailed character, word, and line statistics. Supports Unicode counting and whitespace analysis.

Counts run over Unicode code points, not grapheme clusters: a CJK character is 1, a ZWJ emoji or a flag is several. Reading time assumes 200 words per minute for Latin script.

Tool workspace

0
Characters (graphemes)
0
Characters (no spaces)
0
Words
0
Lines

Structure

Sentences0
Paragraphs0
Bytes (UTF-8)0
UTF-16 code units0

Character Types

Letters0
Digits0
Punctuation0
Spaces0
Other0

Unicode

Unicode code points0
Emoji0
CJK Characters0

Estimates

Reading Time0 min
Speaking Time0 min
Avg Word Length0.0 chars

Unicode Support

The main count uses grapheme clusters — what people perceive as characters. A ZWJ emoji, a flag, or an accented letter counts as 1. The detail panel also shows UTF-16 code units (what String.length returns) and Unicode code points.

Word Boundaries

Words are split on whitespace only, so punctuation does not separate them — hello,world counts as one word — and CJK text without spaces reports a single word (e.g. 你好世界这是我写的文章 → 1 word, 11 CJK characters). Use the CJK Characters figure for those texts.

Reference: a real run, the parameters, and the failure modes

Worked example

A German sentence with a compound word, an abbreviation and a date

Input

text:
Die Datenverarbeitungsvertrag-Prüfung ist z.B. am 31.12.2026 fällig.

Output

words: 7 · characters with spaces: 68 · without spaces: 62 · UTF-8 bytes: 70 · sentences: 5

How to verify: `printf %s 'Die Datenverarbeitungsvertrag-Prüfung ist z.B. am 31.12.2026 fällig.' | wc -w` prints 7 and `wc -m` prints 68. The sentence count is higher than a grammar-aware counter because z.B. and the date add periods.

Parameters

Parameters
ParameterValueDefaultBound
CharactersUnicode code pointscode pointsA ZWJ emoji is several code points but one grapheme; the grapheme panel shows both
WordsRuns separated by whitespacewhitespace splitHyphenated compounds and z.B. count as one word
SentencesSplit on . ! ?—Abbreviations and dates add sentence breaks
Reading timeWords ÷ 200, or ÷ 130 in slow mode200 wpmRounded up to whole minutes
BytesUTF-8 byte length—Umlauts take two bytes, ASCII one
Case toolslower / UPPER / Title—ß becomes SS in uppercase, so the length changes

Failure checklist

Failure checklist
SymptomLikely causeWhat to do
The word count is higher than in Word or Docs.The text contains non-breaking spaces or soft hyphens.Remove invisible characters, then compare with the character panel.
The character count is lower than the editor shows.Most editors count UTF-16 units, so a surrogate pair adds two.Use the UTF-16 panel for the editor-like number.
The sentence count looks too high.Abbreviations (z.B., Dr.) and dates contain periods.Expected: the count is punctuation-based, not grammatical.
The byte count does not match the file size.The file is UTF-16 or starts with a BOM.Save as UTF-8 and compare again.
Uppercase output is longer than the input.ß became SS.Expected for German; re-check any length limit afterwards.
Use cases
  • 1Count characters, words, and lines before posting or sending content with length limits
  • 2Verify Unicode-sensitive text like emoji or CJK content for accurate character counts
  • 3Estimate reading and speaking time for articles, scripts, and documentation
  • 4Check text structure — sentences, paragraphs, and character types — before publishing

Frequently Asked Questions

Which statistics are provided?

Characters (with and without spaces), words, lines, sentences, paragraphs, bytes, and reading time.

Does it handle CJK and emoji correctly?

Yes. Counting is Unicode-aware, so Chinese, Japanese, Korean, and emoji are counted accurately.

Is my text uploaded?

No. All analysis happens locally in your browser.

How are words counted for Chinese, Japanese, and Korean?

CJK text uses a script-aware counting rule instead of splitting on spaces, so a paragraph without spaces still returns a word count. Latin text is split on whitespace and punctuation boundaries.

How is reading time calculated?

Reading time is estimated from the word count at a fixed reading speed and rounded up to whole minutes. It is a planning estimate, not a measurement of how long a particular reader will need.

Related Guides

Related Tools

Your inputs remain on your device.

You are responsible for key management and output validation.