Word Counter

This word counter updates as you type or paste, showing total words, characters, characters without spaces, lines, UTF-8 bytes, and estimated reading and speaking times based on fixed speeds. It is useful for essays, abstracts, blog posts, social media copy, subtitles and application forms with length limits. One thing to know first: 'Total Words' follows this tool's own rules. Chinese characters and Japanese kana are counted one by one, and any other run of text separated by whitespace counts as a 'word' if it contains a letter or digit. Punctuation that stands alone is not counted, as explained below. As a result, the number may differ slightly from Word, WPS or a platform's own count, so when a submission has a strict limit, trust the count from the platform you are submitting to.

How to use it

1. Type in the 'Input Text' box or paste your text. No button is needed, and the statistics update live as you type.
2. Read the seven figures below: 'Total Words', 'Characters', 'Characters (no spaces)', 'Lines', 'Bytes (UTF-8)', 'Estimated Reading Time' and 'Estimated Speaking Time'.
3. To count one passage, paste just that passage. The tool counts everything in the box and cannot count only a selected part.
4. Clear the box and the numbers return to 0, with the time estimates showing 'Less than 1 min'.

How each number is calculated

• Total Words: the number of Chinese characters (Han ideographs including extension blocks) and Japanese kana, each counted as one, plus the number of 'words'. Words are found by replacing every Han character and kana with a space and splitting the rest on whitespace; each piece containing at least one letter or digit counts as one word (punctuation-only pieces are ignored).
• Characters: the length of everything in the box, including spaces and line breaks. Characters that take two UTF-16 code units, such as emoji, count as 2.
• Characters (no spaces): the length after removing all whitespace (spaces, tabs, line breaks).
• Lines: the number of pieces when the text is split on line breaks. Empty text is 0, and a trailing empty line adds one.
• Bytes (UTF-8): the size of the text encoded as UTF-8. A Chinese character usually takes 3 bytes, a Latin letter or half-width punctuation mark takes 1. Handy for checking a database field or API length limit.
• Estimated Reading Time: assumes about 300 Chinese characters or 200 English words per minute. Under one minute shows 'Less than 1 min'; otherwise it rounds up to 'About N min'.
• Estimated Speaking Time: twice the reading time, with the same rounding. These speeds are common rules of thumb, and people vary a lot.

A worked example

Enter two lines. The first is 今天天气很好,我们去公园散步。 (with a half-width comma) and the second is 'Hello world, this is a test.' The tool reports Total Words 19, Characters 44, Characters (no spaces) 38, Lines 2, Bytes 72, and both times as 'Less than 1 min'. The 19 breaks down as 13 Chinese characters plus 6 English words (Hello, world, this, is, a, test, with their trailing punctuation attached). The lone ',' and '。' left in the first line between the Chinese characters are punctuation-only pieces and are not counted.

Supported scope and known limits

• Han ideographs (including extension blocks) and Japanese kana are counted one by one. Korean hangul is not counted per syllable; like English it is counted as whitespace-separated words, which is the usual way to count Korean.
• Punctuation that stands alone as its own piece, such as a comma or full stop sandwiched between Chinese characters, or a free-standing dash, is not counted. Punctuation attached to an English word (as in 'world,') adds nothing extra.
• English words are split on whitespace. 'don't' and 'well-known' are one word each, while 'Hello,world' without a space is also one. A number string or a URL counts as one word each.
• Everything in the box is counted, with no distinction between body text, titles, footnotes and quotations.
• The reading and speaking times are rough estimates with no allowance for genre, difficulty or individual pace. For mixed Chinese and English text the two parts are added.
• There is no file upload and no way to read Word or PDF files directly; copy and paste the text. Extremely long text (hundreds of thousands of characters) may slow typing, depending on your device.

FAQ: Why does my count differ from Word or WPS?

Applications treat punctuation, numbers, English words and full-width versus half-width characters differently. Some count punctuation and some don't, and some count a run of letters and digits as one word while others count characters. This tool's rules are described above. If your submission platform has an explicit counting rule, follow its count.

FAQ: What is the difference between characters and bytes?

Characters are individual symbols; bytes are the storage space after encoding. For mostly Chinese text, the byte count is roughly three times the character count. Check characters when a system limits 'maximum characters', and bytes when the limit is in bytes, as with some database fields.

FAQ: How are the reading times estimated, and are they accurate?

The formula is Chinese characters divided by 300 plus English words divided by 200, giving minutes. It reflects a common average, but real reading depends on difficulty and personal habit, and speaking is also affected by pace and pauses. It helps you judge roughly how long a piece will take, but is not suitable for exact scheduling. For a speech, read it aloud and time yourself.

FAQ: Is my pasted text saved anywhere?

No. The text exists only on the current page and disappears when you reload or close it. It is not sent to any server, so pasting a draft is safe.

Privacy and processing

All counting happens in your browser. The code contains no request that sends your text anywhere and writes nothing to local storage. The byte count is calculated locally using a built-in browser feature.