文字数カウンター

A word counter reads a block of text and reports four numbers that writers, editors, translators, and platform reviewers all care about: how many words, how many characters, how many sentences, and how many paragraphs the text contains. None of those four are as obvious as they sound — they depend on language, on whether spaces and punctuation count, and on what the destination platform considers a word boundary. The tool exists to give a single, predictable answer instead of letting every editor and CMS implement its own rule.

Why one global "word count" cannot work

English-style word counting splits on whitespace: "two words" is two. Chinese, Japanese, and Korean writing have no spaces between words, so the same rule would either return 1 or return the number of characters depending on how the splitter is written. Programmer text adds another wrinkle — is camelCase one word or three? Different tools answer differently, so the more important question is not "which number is right" but "which number am I being asked for".

How each metric is computed

The four metrics are independent. Knowing how each one is defined makes it much easier to predict the output and to explain it to a colleague who asks why "the word count looks wrong".

• Words: contiguous runs of letters or digits separated by whitespace and most punctuation. Hyphenated compounds count as one word; apostrophes inside a word ("don't") do not split it.
• Characters: every visible character, including punctuation. A separate count without spaces is often more useful for platforms that gate input by character limits.
• Sentences: runs of text ended by ., ?, or !. Abbreviations like "Dr." or "e.g." can mislead a naive splitter; for review purposes the count is a reasonable indicator, not a contract.
• Paragraphs: blocks separated by blank lines. A single hard newline does not start a new paragraph — that is a line break, not a paragraph break.
• Reading time: an estimate derived from the word count assuming a typical reading speed (often 200–250 words per minute). It is a hint for content planning, not a measurement of any individual reader.

Rule of thumb: the platform that will display or accept the text always has the final word on counting. Use this tool to draft and budget; verify against the destination platform before treating any number as a hard limit.

Which metric matters when

このツールの使い方

1. Prepare representative pasted articles, abstracts, messages, drafts, and translated copy in 文字数カウンター instead of starting with the largest or most sensitive real input.
2. Run the workflow, generate length metrics that show how large the text is before submission or publication, and review word boundaries, CJK character counts, paragraph breaks, pasted whitespace, and reading-time assumptions before deciding the result is ready.
3. Only copy or download the result after it fits article editing, submission limits, translation estimates, SEO snippets, and billing checks and no longer conflicts with this constraint: Different platforms count words and characters differently, so use the target platform's counter for hard limits.

文字数カウンター の例

この例は、文字数カウンター が想定している入力の形と、自分の作業に使う前に確認しておきたい結果の見え方を示しています。

入力例

Fast browser tools for everyday work.

期待される出力

Words: 6
Characters: 38
Estimated reading time: under 1 minute

A short paragraph and the numbers it produces

Text:
"Welcome to the launch. Doors open at 6 PM.
Bring a printed invitation. Refreshments will be served."

Counted:
  Words:                15
  Characters:           107
  Characters (no space):91
  Sentences:            4
  Paragraphs:           1
  Reading time:         ~5 seconds

Where a quick count actually saves time

Counting before writing further is most useful when there is a hard ceiling, a price tag, or a downstream layout that depends on the answer. The list below covers the cases where checking the number changes the next decision.

• Drafting a tweet, an Instagram caption, or a meta description against a fixed character budget.
• Pricing a translation job by word count and producing an estimate the client can verify.
• Submitting to an academic journal, scholarship form, or job application that caps each section.
• Newsletter or blog scheduling — picking pieces that fit a specific reading-time slot.
• Editing for plain-language guidelines that require shorter sentences and tighter paragraphs.

Edge cases that make the number disagree with intuition

When a count surprises you, it is almost always one of these. Knowing them up front saves a long debate about whether the tool is broken.

• Chinese, Japanese, Korean text has no spaces. A splitter that assumes English will report 0 or 1 words; what you usually want is the character count instead.
• Emojis and combining characters: a single emoji often shows as one glyph but counts as multiple code points. Different platforms then collapse them differently.
• Hyphens and apostrophes: "state-of-the-art" is one word here, four words on many platforms; "isn't" is one word here, two on some character-strict counters.
• Numbers and code: "v2.0.1" is one token here, may be three sentences on a naive sentence splitter because of the dots.
• Markdown and HTML formatting: do you want to count the markup characters or the rendered text? Most counters count the source as written, not what the reader sees.
• Inserted whitespace: copy from rich-text editors often carries non-breaking spaces, zero-width joiners, and trailing line breaks that visibly add to character counts.

How different platforms count, and how this tool aligns

実用上の注意

• 文字数カウンター は既定でブラウザ内で動作するため、別のツールチェーンを用意せずにすばやくローカル確認を行えます。
• 実際の入力が大きい、機密性が高い、または業務上重要な場合は、まず代表的なサンプルから始めてください。
• 本番環境、顧客向け、法務、財務、安全性が重要な作業に使う前に、最終結果を必ず確認してください。

文字数カウンター の参考情報

文字数カウンター は、入力の整理、繰り返し可能な変換、公開向け出力を説明します。

• 長いテキストを処理する前に、空白、改行、句読点、見えない文字を確認してください。
• 重要な文章を置換、並べ替え、重複除去、比較する場合は、まず小さなサンプルで試してください。
• 生成された slug、HTML、比較結果は公開前に確認してください。

参考資料

• Unicode Standard Annex #29: Text Segmentation
• MDN: Intl.Segmenter