How to Remove Unicode Characters from Text
To remove unicode characters from text, paste it into the tool and read the result on the right. Unicode is the standard that gives every letter, symbol and emoji in every language its own number. ASCII is the small part of it that holds the 128 characters of an old English keyboard. Anything outside those 128 is what this tool removes.
Here is the sample text with the default settings. Accents and curly quotes are converted, and what has no plain equivalent is removed:
Input
Café Zoë wrote: “It’s a naïve plan – Straße & crème brûlée…”
Price: 5€, 2½ kg, fine print ™
Tokyo: 東京 🚀 A1
Output
Cafe Zoe wrote: "It's a naive plan - Strasse & creme brulee..."
Price: 5, 2 1/2 kg, fine print TM
Tokyo: A1Four characters were removed (the euro sign, the two Japanese characters and the rocket) and 17 were converted. The list under the result names every one of them.
Remove Non-ASCII Characters and Keep the Rest
The default mode, Non-ASCII characters, is the one to use when a system rejects anything outside plain ASCII. It leaves letters, digits, punctuation, spaces and line breaks exactly as they were. Only characters with a code above 127 are touched.
That includes characters you cannot see. A non-breaking space, a zero-width space and a byte order mark are all outside ASCII, so they are handled too. The non-breaking space becomes a normal space when Convert to ASCII first is on. For a closer look at invisible characters, use the Invisible Character Remover.
Convert to ASCII First: What Becomes What
Deleting every accented letter makes words hard to read: with Convert to ASCII turned off, the Café of the sample text becomes Caf. So conversion runs first. A letter with an accent loses the accent, and a few letters have a set spelling. Curly quotes, dashes and the ellipsis become their plain versions.
- Accents: é è ê ë ï ñ ü and similar become e e e e i n u. Combining accent marks on their own are dropped.
- Special letters: ß becomes ss, æ becomes ae, œ becomes oe, ø becomes o, þ becomes th, ł becomes l.
- Quotes: curly single quotes become ', curly double quotes and « » become ".
- Dashes and spaces: en and em dashes and the minus sign become -, and non-breaking and other wide spaces become a normal space.
- Look-alikes: fullwidth letters such as A1 become A1, ligatures such as fi become fi, circled digits become digits, ™ becomes TM and © becomes (c).
- Fractions: ½ becomes 1/2, with a space added after a digit so 2½ becomes 2 1/2.
Letters from other alphabets, such as Greek, Cyrillic, Arabic, Hebrew and Chinese, have no plain equivalent here, so they are removed. The tool does not transliterate them into Latin letters. All dashes become a single hyphen, so an em dash and an en dash look alike afterwards.
Remove Special Characters from Text
To remove special characters from text, set Remove to All special characters. This keeps only the letters A to Z, the digits 0 to 9 and spaces. Every other character is removed, including punctuation such as commas and the & sign. Line breaks and tabs stay when Keep line breaks and tabs is on.
With that mode, Hello, World! (2026) #1 becomes Hello World 2026 1. Accented letters are converted first when conversion is on, so Café becomes Cafe and not Caf. Only letters and digits survive the conversion in this mode, so © becomes c and ½ becomes 12 instead of (c) and 1/2. If you only want to remove punctuation and keep letters of every language, use Remove Punctuation instead.
Remove Accents from Text
To remove accents from text, leave the mode on Non-ASCII characters and Convert to ASCII first on. Every accent is stripped from its letter and the rest of your text is kept as it was. Names such as José Núñez become Jose Nunez.
The same step also straightens curly quotes and replaces dashes, because those are not ASCII either. If you want to keep a character such as an em dash, this tool is the wrong one for that text, since it works on the whole text at once.
Why Strange Characters Appear in Text
Text copied from a word processor, a web page or a chat app often carries characters that look like normal ones. Curly quotes, long dashes, non-breaking spaces and the three-dot ellipsis are the most common. They are valid Unicode, but they break a CSV import, a SQL script, a filename, an SMS or a form that expects plain ASCII.
The list of changed characters tells you which ones were there. A line such as ’ → ' x4 means four curly apostrophes were found. If your text already shows strings such as é in place of é, the text was decoded with the wrong encoding before it reached you. This tool cannot repair that. It would turn é into A(c).
Remove Non-ASCII Characters in Python or JavaScript
In code, deleting non-ASCII characters is one line. Both of these only delete. They do not convert é to e like this tool does:
text = "Café ☕"
print(text.encode("ascii", "ignore").decode())const text = "Café ☕";
console.log(text.replace(/[^\x00-\x7F]/g, ""));Both print Caf with a trailing space where the ☕ was. For a one-off file or a quick check of what is in the text, pasting it into the tool is faster than writing a script.
Limits and When to Use Something Else
- Emojis are removed in both modes. To clean only emojis and keep accents and other characters, use Remove Emojis.
- To find zero-width and other invisible characters before deciding, use the Invisible Character Remover.
- Characters are treated one by one. A flag emoji, or a letter with several marks, counts as more than one character in the list.
- The tool changes text only. It does not re-encode a file between Latin-1, Windows-1252 and UTF-8.
Privacy
Your text is processed in your browser. Nothing is uploaded or stored. See the Privacy Policy for details.