URL Slug Generator

How to use

1. Enter article titles or page names in “Titles (one per line)”. Paste several lines to get one slug per line. “Insert example” fills in an example
2. Under “Characters”, choose a slug of letters and digits only, or one that keeps Japanese. The first setting is “Letters and digits only (kana as romaji)”
3. If needed, choose how to write long vowels, the separator (hyphen or underscore), the letter case, and a maximum length. Enter something like https://example.com/blog/ in “Text before the slug” to copy the slugs as URLs too (the URL of the first slug is shown as a sample)
4. Use the checkboxes to write symbols as words, write ä ö ü the German way, number lines that give the same slug, or remove English stop words
5. The slugs appear in “Slugs (one per line)”. “Copy slugs” puts them on the clipboard. When you entered the text before the slug, “Copy URLs” is there too

There is no run button: the result changes as you type or change a setting. The result has one line for each line of input (a blank line stays blank), so it lines up when pasted into a spreadsheet column. Everything you enter is processed in your browser and is never sent or saved anywhere.

Below the result are notes about removed kanji, kana read as one long vowel and numbered lines, and a table with each line’s slug and its length.

Example

The five lines that “Insert example” fills in, with the first settings:

How to Brew Coffee at Home   → how-to-brew-coffee-at-home
10 Tips for Better Sleep!    → 10-tips-for-better-sleep
Crème brûlée & Café          → creme-brulee-and-cafe
Q&A: What's New in 2026?     → q-and-a-whats-new-in-2026
How to Brew Coffee at Home   → how-to-brew-coffee-at-home-2

• Letters are lowercased, and spaces and punctuation become a single -
• The accents on é and û are removed, and & becomes and
• The apostrophe in What's is removed without splitting the word
• Line 5 gives the same slug as line 1, so -2 is added at the end

With “Remove English stop words” on, the same lines become:

how-brew-coffee-home
10-tips-better-sleep
creme-brulee-cafe
q-whats-new-2026
how-brew-coffee-home-2

Japanese titles work the same way. はじめての プログラミング becomes hajimeteno-puroguramingu, and Next.js 16の新機能 becomes next-js-16-no with a note that the kanji 新機能 were removed. Written with its reading as Next.js 16の しんきのう, it becomes next-js-16-no-shinkino. With “Characters” set to “Keep Japanese”, it becomes next-js-16の新機能, and the table’s “Encoded” column shows its length once percent-encoded for a URL, where each kana or kanji takes 9 characters (such as %E3%81%AF).

How text is converted

Each line goes through these steps:

• Fullwidth letters and digits become ASCII, halfwidth katakana become fullwidth, and ① and ² become 1 and 2 (Unicode NFKC normalization)
• Commas every three digits, as in 1,000, are removed to give 1000. The point in 1.5 is a separator, giving 1-5
• With “Write symbols as words” on, & becomes and, + becomes plus, @ becomes at, and % becomes percent. C++ becomes c-plus-plus
• Accents are removed from letters (é → e, ñ → n). ß becomes ss, æ becomes ae, ø becomes o, ł becomes l, and þ becomes th. The German style writes ä ö ü as ae oe ue
• An apostrophe inside a word is removed without splitting it (Don't → dont)
• Anything other than letters, digits and kana (spaces, punctuation, brackets, symbols, emoji) is a separator. Separators in a row become one, and separators at the start and end are removed
• With “Letters and digits only”, the point where Latin letters meet kana is also a separator (iPhoneケース → iphone-kesu, Tシャツ → t-shatsu)
• “Lowercase” lowercases the letters. Romaji is always lowercase

Kana as romaji

Kana are written in Hepburn romaji: し is shi, ち is chi, つ is tsu, ふ is fu, じ and ぢ are ji, ず and づ are zu, and を is o.

• ん is always n (しんぶん → shinbun, しんばし → shinbashi)
• っ doubles the next consonant (がっこう → gakko), and is t before ち (まっちゃ → matcha)
• Loanword sounds such as ティ ファ ウェ ヴァ take their common spellings (ti fa we va)
• Long vowels “Leave out” writes おう, おお and うう as one vowel and leaves out ー (とうきょう → tokyo, おおさか → osaka, コーヒー → kohi). As in passport names, えい, いい and ああ are written out (せんせい → sensei)
• Long vowels “As the kana spell them” writes every kana, and ー repeats the vowel before it (toukyou, koohii)

With “Leave out”, the もう in おもう (to think) is also read as a long vowel and gives omo. When the kana are separate sounds, type ' between them, as in おも'う, to get omou. The note below the result lists which kana were read as long vowels.

は and へ are written as spelled, ha and he. To write the particle は as wa, type わ.

Choosing a slug

• Google recommends readable words rather than long ID numbers in URLs, and hyphens (-) rather than underscores (_) between words
• Google Search treats URLs as case sensitive: /Apple and /apple are different URLs. If your server returns the same page for both, pick one case
• A slug that keeps Japanese shows as Japanese in the address bar, but when copied or shared it turns into percent-encoding such as %E3%81%AF and gets long
• WordPress stores a Japanese slug percent-encoded and cuts anything past 200 encoded characters (about 22 kana or kanji). A slug of letters and digits is also kept to 200 characters. Rows over 200 are listed in the note below the result and in the table’s “Notes” column
• Changing the slug of a published page breaks links to the old URL. If you change it, set up a redirect

FAQ

Can it turn kanji into romaji?

No. The reading of a kanji depends on context (今日 can be kyō or konnichi) and cannot be chosen without a dictionary. Kanji are removed and listed in a note, so rewrite those lines with the reading in hiragana or katakana, or in English. To keep the Japanese, set “Characters” to “Keep Japanese”.

Removing the kanji left a meaningless word

Only the kana between the kanji become romaji, like the の in Next.js 16の新機能. The table’s “Notes” column shows the removed characters for each line, so replace those lines with ones that spell the reading in kana.

Where do Japanese kana words break?

Only where you type a space. Japanese is written without spaces between words, so とうきょうのてんき becomes the single word tokyonotenki. Typed as とうきょう の てんき, it becomes tokyo-no-tenki.

Which is better: letters and digits only, or keeping Japanese?

A Japanese slug makes the page easier to recognize in search results and the address bar, but turns into a long percent-encoded string when the URL is shared on social media or by email. Google can handle URLs in languages other than English. It helps to decide by how often the URL will be shared.

What happens to lines that give the same slug?

With “Number lines that give the same slug” on, later lines get -2, -3 and so on (the same form WordPress uses for posts with the same slug). If another line already gives a slug ending in -2, that number is skipped. Slugs that differ only in letter case count as the same. With it off, no number is added and the table’s “Notes” column says “Same as line …”.

How is the maximum length counted?

In characters of the slug (with “Keep Japanese”, each kana or kanji is one character). Words are not cut in half: the slug ends at the last separator within the limit. A single word longer than the limit is cut where the limit falls. A numbered slug fits within the limit, number included.

What are stop words?

Words such as a, the and of that appear in almost every English sentence and carry little meaning. They are sometimes removed to keep slugs short. You can edit the list of words to remove. The list is read the same way as titles, so für removes the word fur and & removes and. An entry that splits into two words, such as e-mail, cannot be used. Only words typed in Latin letters are removed, so the romaji to (と) and no (の) stay. A line made only of stop words keeps all of them.

Is what I enter sent anywhere?

No. Everything is converted in your browser, and nothing you enter or get is sent to a server or saved. It is gone when you close the page.