Word Cloud Generator
Turn any text into a word cloud sized by how often each word appears.
About this tool
Turns a block of text into a word cloud, where a word is drawn larger the more often it appears. Paste an article, a transcript, survey responses, or your own draft to see at a glance which terms dominate. The image is drawn on a canvas in your browser and can be downloaded as a PNG — nothing is uploaded.
How it's built. The text is split into words, lowercased, and counted. Optionally a list of ~80 common English "stop words" (the, and, of, to, is, it, that…) is removed so the cloud shows content rather than grammar. The remaining words are ranked by frequency, the font size is scaled to the count, and words are placed largest-first, spiralling outward from the centre to find open space.
Why some words don't appear. Placement is greedy: once the canvas fills, a smaller word that can't find a gap is skipped rather than overlapped. If words you expected are missing, lower "Max words", shorten the text, or accept that the least frequent terms won't fit — they're the least informative anyway.
What a word cloud is good for. A fast, visual gist of a document; spotting repeated themes in open-ended survey answers or reviews; a title graphic for a talk or blog post; comparing two texts side by side. It's an overview, not analysis.
Its limitations. Frequency isn't importance — a word repeated for emphasis outweighs a key term used once. Word clouds ignore order, negation ("not good" reads as "good"), and context, and they make near-synonyms ("car", "cars", "vehicle") look separate. For anything you'll draw conclusions from, pair it with the actual counts.
Getting a cleaner cloud. Edit the source text before pasting: remove boilerplate (headers, navigation, signatures), and if a proper noun or phrase is splitting oddly, join it with an underscore so it counts as one token. There's no custom stop-word list, so unwanted words are best removed at the source.
For the underlying numbers, use the word frequency counter or keyword density checker. For a plain total, the word counter.
Frequently asked questions
- Why did some words get left out?
- Words are placed largest-first with a spiral search for space. If the canvas fills before a smaller word finds a gap, it's skipped rather than overlapped. Lower "Max words" or shorten the text to fit more.
- Can I control which words are excluded?
- The "Ignore common words" toggle filters a built-in list of ~80 English filler words. There's no custom exclude list — remove unwanted words from the text before pasting.
- Does word size mean importance?
- It means frequency, which isn't the same thing. A word repeated for style can outweigh a crucial term mentioned once. Read the cloud as a gist, not a ranking of significance.
- Why are "run", "runs", and "running" counted separately?
- The tool counts exact word forms and doesn't stem or lemmatise. Related forms and synonyms show as distinct entries.
- Is my text uploaded anywhere?
- No. Counting and rendering both happen entirely in your browser.
- Can I use this for another language?
- The layout works for any space-separated language, but the stop-word filter is English only — turn it off, or pre-clean the text, for other languages.