読書時間推定ツール

How reading time is actually calculated

Reading time is a deceptively simple idea: count the words, divide by a reading speed measured in words per minute (WPM), and you get minutes. The honest part is admitting that no single speed fits everyone, which is why this tool reports several. It counts the words in your text and divides by four different speeds at once, so you can pick the one that matches your audience:

It also gives a speaking time at 130 WPM, because people speak far more slowly than they read silently — a useful figure for scripts, presentations, and voice-overs.

A worked example

Paste an 1,800-word article. At the average 200 WPM, the estimate is 1800 ÷ 200 = 9 minutes. A careful reader at 150 WPM needs 1800 ÷ 150 = 12 minutes, while a fast reader at 250 WPM finishes in about 7.2 minutes, shown as 7m 12s. For a spoken delivery at 130 WPM the same text runs roughly 13.8 minutes — which is why a blog post that reads in nine minutes becomes a much longer talk. Seeing all of these at once stops you from quoting a single misleadingly precise number.

The readability score, and what it means

Beyond time, the tool computes the Flesch Reading Ease score, a long-established readability formula. It uses three measurable quantities — total words, total sentences, and total syllables — in this relationship:

Ease = 206.835 − 1.015 × (words ÷ sentences) − 84.6 × (syllables ÷ words)

In plain terms, longer sentences and longer words both push the score down. A higher score means easier reading. Roughly: 90–100 is very easy (around 5th-grade level), 60–70 is plain English most adults read comfortably, 30–50 is difficult and reads like academic prose, and below 30 is very hard. The tool also reports the related Flesch-Kincaid Grade Level, which translates the same inputs into a U.S. school grade — a grade of 8 means a typical 13- to 14-year-old should follow it.

How syllables are estimated

English syllable counting has no perfect rule, so the tool uses a fast heuristic: it lowercases each word, removes common silent endings (like a trailing e or ed), and counts groups of vowels, treating consecutive vowels as one syllable. This matches a human count most of the time but can misjudge unusual words. Because the readability formula averages over your whole text, a few miscounted words barely move the score — but for a single short sentence the estimate is rougher.

Practical uses

• Setting expectations on articles. A "5 min read" label up front improves how readers engage with a post.
• Timing talks and videos. Use the speaking-time figure to fit a script into a slot before you rehearse.
• Tuning content for an audience. If a help article scores 35 on the ease scale, you know to shorten sentences and swap jargon for plain words.
• Editing for clarity. The average sentence length and longest-words list quickly highlight where prose has become dense.

Tips and common mistakes

• Don't treat the minutes as exact. They are estimates; comprehension speed varies with topic, formatting, and how many images or code blocks break up the flow. Round to a friendly number.
• Readability formulas measure form, not sense. A high ease score does not guarantee the writing is clear or correct — short, simple sentences can still be vague. Use the score as a prompt to review, not a verdict.
• The formula is tuned for English. Applying Flesch scores to other languages, or to text full of code, URLs, or numbers, gives unreliable results because the syllable model assumes English words.
• Sentence detection relies on punctuation. Text without periods, or with many abbreviations, can throw off the words-per-sentence figure and therefore the score.

Privacy

All counting and scoring happens in your browser as you type — the text is analyzed locally and never uploaded or stored. You can paste a confidential draft and it stays on your device, working even with no internet connection.