Remove punctuation from text

Plain words, without the punctuation

Punctuation is essential for readers, but it gets in the way of many text tasks: counting how often each word appears, comparing two texts, preparing data for analysis, making a list of words from a sentence, matching song lyrics or checking a text against a list. In all of those, "end." and end should count as the same word. Removing punctuation first makes the rest of the work simpler.

What counts as punctuation

The tool removes every character that Unicode classifies as punctuation. That covers full stops, commas, colons, semicolons, question and exclamation marks (including the Spanish inverted ones), quotation marks of all styles, brackets and parentheses, dashes, hyphens, apostrophes, slashes and ellipses. Punctuation from other writing systems is included too, such as the Chinese full stop or the Arabic comma.

Keep what belongs to the word

Deleting every mark would turn don't into dont and well-known into wellknown, and 3.5 into 35, which changes the meaning. Two options prevent that: one keeps apostrophes and hyphens when they sit between two letters, and the other keeps a point or comma between two digits. Both are on by default, so only the marks that separate words and sentences disappear.

Symbols

Symbols such as $, €, +, =, @, # and © are not punctuation in the technical sense, so they are kept unless you tick "Also remove symbols". Use that option when you want only letters, digits and spaces.

Delete or replace with a space

By default each mark is deleted. If words are joined only by punctuation, as in one,two,three or and/or, deleting would merge them into a single string. Choose "Replace each mark with a space" to keep the words apart; repeated spaces created that way are collapsed.

What it does not do

The tool does not understand the text, so it cannot tell that a full stop in e.g. or in a web address is part of the expression. Check the result if your text contains abbreviations, URLs or email addresses.