Fake Word & Definition Generator
Generates made-up words with plausible-sounding dictionary-style definitions
About this generator
A fake word and definition generator gives you invented terms that read like they escaped from a real dictionary — useful for Balderdash-style party games, fictional glossaries, worldbuilding notes, and dictionary UI mockups. Each entry pairs a made-up headword with a part-of-speech label and a one-line definition, formatted like "Blorention (noun): The act of quietly observing trivial domestic rituals." Under the hood, words are assembled from 15 prefixes, an optional connecting syllable, and 10 suffixes like -tion, -ism, and -ify, which is why they sound plausibly English. Definitions combine one of 8 openers, 8 actions, and 8 objects — 512 possible sentences — and each word is randomly tagged noun, verb, adjective, or adverb. Set the count anywhere from 1 to 20 entries per run. One honest caveat: the part-of-speech tag is chosen independently of the definition, so you will sometimes see a "verb" defined as "Relating to…". For polished use — a printed glossary, say — give the labels a quick human pass.
How to use
- Set the count field to the number of fake words you want, between 1 and 20.
- Click Generate to produce a batch of invented words with dictionary-style definitions.
- Read through the results and copy any entries that fit your project or game.
- Regenerate as many times as needed to get a variety of tones, parts of speech, and meanings.
- Paste selected definitions directly into your document, prototype, or game materials.
Use Cases
- •Populating the glossary appendix of a fantasy novel with 10–15 coined terms and their in-world meanings
- •Generating spell, creature, and location names mid-session for a D&D or Pathfinder campaign with definitions ready to read aloud
- •Filling a Figma dictionary-app mockup with realistic headwords and definitions so stakeholders can evaluate line height and truncation
- •Running a classroom etymology exercise where students reverse-engineer plausible Latin or Greek roots from invented words
- •Building a round of Balderdash or a custom party game that needs 8–10 obscure-sounding words with convincing fake definitions
Tips
- →Generate in batches of ten or more and cherry-pick: the best three from ten beats forcing one from one.
- →Look for entries where the definition suggests an emotion or social situation — those work best in fiction and party games.
- →If you need all nouns or all verbs for consistency, keep regenerating and filter by the part-of-speech label in each result.
- →Pair a strong fake word with a real etymology reference: if the generated word starts with 'morb-', research what Latin roots that prefix evokes to deepen your worldbuilding.
- →For Balderdash-style games, avoid words that are too short or too phonetically simple — players find them easier to guess. Longer, unusual-sounding results make better prompts.
- →Save your favorite outputs in a running document; fake words you don't use today often fit perfectly into a project six months later.
FAQ
how are the fake words and definitions actually built
Words are assembled from 15 prefixes, an optional connecting syllable, and 10 suffixes like -tion, -ism, and -ous. Definitions combine one of 8 openers, 8 actions, and 8 objects into a single sentence, and each entry gets a randomly chosen part-of-speech label. That template structure is what makes the entries read like dictionary lines rather than random letters.
why does the part-of-speech label sometimes contradict the definition
The label is picked at random, independently of both the word's suffix and the definition template, so you can get a 'verb' defined as 'Relating to…' or an '-ism' word tagged adjective. For anything polished — a printed glossary or a game handout — give the labels a quick manual pass. It takes seconds and makes the entries hold up to a closer read.
how do I make fake words feel believable in a novel or story
Use the generated definition internally as your guide, but let the word appear in context without explanation — readers infer meaning from action and reaction the same way they do with real unfamiliar vocabulary. Repeat the word at least twice across a chapter so readers learn it organically rather than feeling lectured.
will I get duplicates if I generate the maximum of 20 entries
Entries are drawn independently, so duplicates are possible but rare — there are roughly 1,650 possible word shapes and 2,048 definition sentences. If a batch does repeat a word, one regeneration almost always clears it.
You might also like
Popular tools from other categories that share themes with this one.