Paste a block of text and it segments the words, counts frequencies and packs a word cloud automatically, with six outline shapes, eight palettes and high-resolution PNG export
| Word | Count | State |
|---|
The segmenter built into your browser splits the text for you, so English, Chinese, Japanese and mixed-language passages all work straight away. No need to tidy your keywords into a list first.
Every word is counted, then frequent words are scaled up and rare ones scaled down, so the emphasis of the passage is visible at a glance.
Rectangle, circle, heart, star and triangle, plus a text outline: type one to four characters and the cloud is packed into that lettering instead.
Ocean, forest, sunset, berry and more. Color by frequency, pick colors at random, go monochrome, or paste in your own hex list.
An Archimedean spiral hunts for free space pixel by pixel, so words automatically dodge each other and never end up stacked and unreadable.
Export at 1x, 2x or 3x, up to 3000px wide. Text is redrawn as vectors so the edges stay sharp, and you can copy the image straight to the clipboard.
Drop an article, a set of reviews, meeting notes or a keyword list into the box. You can also click Sample text to try it out, or import a local .txt file.
Look at the word ranking on the right. If a term came out chopped up, add it to the custom dictionary. If a word should not appear at all, add it to the extra stop words or simply click its row to block it.
Choose an outline and canvas ratio, then a palette and font. Nudge the word spacing to let the layout breathe, and hit Shuffle whenever you want a fresh arrangement.
Set the export scale and click Export PNG to download, or click Copy image to paste it straight into a document, a chat window or a design tool.
Turn a long article, report or speech into a cloud and see the themes at a glance
Pool reviews, open-ended survey answers and support tickets to spot the loudest requests
Summarize minutes or lesson points as a cloud; it reads faster than a bullet list on a slide
Turn skills and project history into a personal skill cloud for a CV or portfolio page
Use a text outline to build a themed graphic, then export it as a cover or poster asset
Give a story a word cloud that shows topic heat and word choice at a glance
A word cloud (also called a tag cloud) is a way of visualizing text by arranging its words according to how often they appear. Frequent words get a larger font and a more prominent spot; rare words shrink and fill the gaps. It started out as a quick way to survey the topic of a document and is now widely used for report illustrations, covers and classroom slides.
First each word is counted, then the count is mapped onto the range between the minimum and maximum font size. This tool offers three mappings: square root (the default, which compresses the gap so small words stay legible), linear (the most literal contrast) and logarithmic (which widens the gap further, useful for long texts with a steep power-law distribution).
English words are separated by spaces, so splitting them needs almost no algorithm. Chinese is a continuous run of characters with no delimiters, so deciding whether a phrase breaks as one term or three requires a dictionary and a statistical model. This tool delegates that to the segmenter built into your browser, and you can force merges for new terms or proper nouns through the custom dictionary.
Stop words are terms that appear constantly but carry almost no information: the, a, and, of, or in Chinese 的, 了, 是, 在. Left unfiltered they will almost certainly take the largest slots in the cloud and push the meaningful words into the corners. This tool ships with a built-in stop word list for English and Chinese, enabled by default, and you can turn it off or add your own terms.
A word cloud reflects how often a word appears, not how important it is. A term may repeat simply because it is generic, while the real conclusion shows up only once. It also discards word order and the relationships between words, so it cannot express cause or contrast. Treat it as a first impression and a layout asset; serious analysis still needs the original text and its context.
| Segmentation mode | How it splits text | Best for |
|---|---|---|
| Smart | Uses the dictionary built into the browser and splits by meaning | The vast majority of cases; the default and first choice |
| Bigram | Splits a run of CJK characters into adjacent pairs | Texts full of new terms and proper nouns; expect a little cross-word noise |
| Single character | Every CJK character stands alone; Latin words stay whole | Character frequency studies, calligraphy or glyph themes |
| Latin words | Takes only Latin-alphabet words and ignores CJK text | Purely English text, or when you want to look at English wording on its own |
| Outline | Visual character | Good for |
|---|---|---|
| Rectangle | The largest usable area, so it fits the most words | Information density first; formal report illustrations |
| Circle | Gathers toward the middle and tapers at the edge | Badges, avatars and square social posts |
| Heart | Instantly recognizable and full of mood | Holiday greetings and emotional themes |
| Star | Sharp angles with strong visual impact | Event posters, children's and school themes |
| Triangle | Narrow on top, wide at the base, clearly layered | Expressing hierarchy or progression |
| Text shape | The cloud itself forms the outline of a character or word | Theme words, brand names and title covers |