Skip to content

Vocabulary Analyzer

Measure vocabulary richness: type-token ratio (plus root and corrected TTR), unique words, words used once, lexical density and word-length distribution.

0 characters 0 words

Related Tools

Word Counter Readability Checker Word Frequency Counter Text Statistics

How to Use

  1. Input Your Text — Paste your essay, article, story, or any text sample into the text area. For reliable vocabulary analysis, use at least 200 words to ensure a meaningful sample size for diversity calculations.
  2. Run Vocabulary Analysis — Click Analyze Vocabulary to calculate lexical diversity metrics. The tool identifies unique words, measures type-token ratio, root TTR and corrected TTR, and reports lexical density and the share of long words.
  3. Interpret Diversity Metrics — Review your vocabulary profile including unique word count, type-token ratio, hapax legomena (words used only once, listed so you can see them), a word-length histogram and your most repeated content words. Copy the report or use these insights to diversify or refine your word choices.

About the Vocabulary Analyzer

The Vocabulary Analyzer measures how varied your word choice is. Paste a draft and it reports how many different words you used, how much you repeat yourself, which content words you lean on most, and how long your words tend to be. Students checking an essay, teachers comparing a learner’s drafts across a term, and editors hunting for a pet word all get the same set of numbers, plus a plain-text report they can copy.

How the numbers are calculated

Your text is split into word tokens: runs of letters and digits, lowercased, with internal apostrophes and hyphens kept, so don’t and well-known are one token each. Curly apostrophes count as straight ones, and Chinese, Japanese and Thai text is segmented into words. These are the same tokens the Word Counter and Text Statistics use for unique words, so the three tools agree. The “Words” figure is the ordinary whitespace word count; it can differ slightly from the token count because a number such as 3.14 is one word but two tokens (3 and 14).

MetricHow it is calculated
Type-token ratio (TTR)Unique tokens ÷ total tokens, shown as a decimal and a percentage
Root TTR (Guiraud)Unique tokens ÷ √(total tokens)
Corrected TTR (Carroll)Unique tokens ÷ √(2 × total tokens)
Words used once (hapax legomena)Tokens that occur exactly once, with their share of unique words; the first 80 are listed on screen and the copied report lists all of them
Average word lengthCharacters per token, counting every character in the token, including an internal apostrophe or hyphen
Lexical densityShare of tokens that are not on the built-in list of 161 common English function words (the, of, was, however and so on)
Long wordsShare of tokens with 7 or more characters
Most repeated content wordsUp to 10 words that are not on the function-word list and appear more than once

The results also include a word-length histogram (1 to 15+ characters). A live strip above the results updates TTR, unique words, word count and average length as you type, and a note appears when the sample has fewer than 200 tokens.

Worked example

Input: The cat sat on the mat. The cat was happy, and the mat was warm. Our neighbor’s cat, however, stayed inside all afternoon.

“The” is actually the most frequent token here (4 times), but it is a function word, so it does not appear in the content-word list. The short-sample note is shown because 23 tokens is well under 200.

Limitations

Practical tips

Privacy: every metric is computed by JavaScript in your browser, and the text you paste is not uploaded anywhere.

Frequently Asked Questions

What is type-token ratio (TTR)?

Type-token ratio is the number of unique words (types) divided by the total number of words (tokens) in a text. A TTR of 0.60 means 60 percent of the words in the text are unique. Higher TTR indicates greater vocabulary diversity. However, TTR naturally decreases as text length increases, so comparisons should be made between texts of similar length.

What is a good vocabulary richness score?

There is no single good score. TTR falls as a text gets longer, so a 150-word paragraph and a 3,000-word essay cannot be compared directly, and genre matters too: technical and legal writing repeats precise terms on purpose, while descriptive fiction tends to vary its words more. The most useful comparison is against your own earlier writing, or a reference text of similar length and genre.

What are hapax legomena and why do they matter?

Hapax legomena are words that appear exactly once in a text. They indicate vocabulary breadth — a high proportion of hapax legomena suggests the writer draws from a wide vocabulary rather than relying on a small set of repeated words. In literary analysis, hapax legomena ratios help distinguish writing styles and assess vocabulary sophistication.

How can I improve my vocabulary diversity?

To increase vocabulary diversity, use a thesaurus to find synonyms for frequently repeated words, expand your reading to expose yourself to varied vocabulary, practice writing exercises focused on word variety, and use our Word Frequency Counter to identify your most over-used terms. Aim to reduce repetition while maintaining clarity and natural tone.

Does vocabulary complexity affect readability?

Yes, vocabulary complexity and readability are closely related but not identical. A text can have high vocabulary diversity (many different words) while still being readable if those words are common and familiar. Conversely, a text with low diversity but many specialized technical terms can be very difficult to read. Our tool measures diversity separately from difficulty.

Can this tool compare vocabulary across different texts?

You can analyze each text individually and compare the resulting metrics. For fair comparisons, use text samples of similar length since type-token ratio is length-sensitive. The type-token ratio, root TTR, corrected TTR and hapax legomena share provide comparable metrics; root TTR and corrected TTR are less sensitive to length, but similar-sized samples still give the fairest comparison.