URL slug generator
Type the title of your page or product and get a clean URL part. Arabic mode keeps Arabic letters as they are (a perfectly valid URL that the browser percent-encodes for you), while Latin mode produces ASCII only. The tool makes no claim of accurate Arabic transliteration and tells you exactly what it dropped. Use it as a URL slug generator for blog posts, product pages and categories.
Loading the tool…
Processing and privacy: The result is calculated inside your browser and your data is not sent to any server.
How to use it
- Type or paste your title, or press “Load example”.
- Choose the mode: “Arabic Unicode” keeps Arabic letters, “Latin ASCII” outputs only a-z and 0-9.
- Set the separator, the case and a maximum length, and turn on the normalization options if you want them (they change spelling).
- Check the slug and its encoded form, edit it by hand if you like, then copy or download it.
Example
The built-in example is one title: “منتج iPhone 15 Pro — قهوة عربية أصيلة & شاي، 2024”.
Arabic mode with the defaults gives “منتج-iphone-15-pro-قهوة-عربية-أصيلة-شاي-2024”, 44 characters long, whose encoded form is 149 characters because every Arabic letter becomes six characters such as %D9%85. Turn all three normalization options on and it becomes “منتج-iphone-15-pro-قهوه-عربيه-اصيله-شاي-2024”.
Latin mode with the defaults gives “iphone-15-pro-and-2024” and lists the 21 Arabic letters that were dropped. Turn the rough transliteration table on and it becomes “mntj-iphone-15-pro-qhwh-arbyh-asylh-and-shay-2024”, which is unvowelled, non-standard and needs your review.
Limits and caveats
- Tested with 100,000 titles in batch mode (about 6.5 MB): the result is computed in about two seconds in Arabic mode and under four in Latin mode with transliteration, but the browser text box needs several seconds to display a paste of that size. Automatic processing therefore stops above about 4,000 lines or 150,000 characters and the “Generate” button takes over. The preview shows only the first 300 lines; copy and download use the full result.
- There is no reliable Arabic-to-Latin transliteration without short vowels: “مكتب”, “مكتبة” and “كتب” are written with almost the same consonants. The table used here is rough, unvowelled and off by default: it maps both ص and س to “s”, both ض and د to “d”, and drops the hamza, so two different words can come out identical. Review and edit the result yourself; that is why the result field stays editable.
- Latin mode strips accents by NFKD decomposition (é becomes e) and handles the letters Unicode cannot decompose through an explicit table: ß to ss, ø to o, ł to l, æ to ae, œ to oe, đ to d, þ to th, ð to d. Anything else outside a-z and 0-9 is removed and listed in the dropped-characters warning.
- The three normalization options (alef variants, ta marbuta to ha, alef maksura to ya) are off by default because they change spelling: “قهوة” becomes “قهوه” and “مستشفى” becomes “مستشفي”. Turn them on if you want one slug for every way of writing a word, and remember that changing a published URL needs a 301 redirect.
- Stop-word removal uses a fixed list of 45 words (21 English and 24 Arabic) and is off by default. The tool does not check whether the slug is already in use and never adds a date or a number to make it unique.
Questions about this tool
Is an Arabic URL really valid?
Yes. RFC 3987 allows non-Latin characters in URLs, every modern browser displays them in Arabic and percent-encodes them on the wire automatically. The only real effect is length: one Arabic letter becomes six characters such as %D9%85, so a 44-character slug is 149 characters in server logs and when copied from some older apps. That is why the tool always shows the encoded form and its length.
Why does the tool not transliterate Arabic into Latin automatically?
Because Arabic is written without short vowels, so no program without a dictionary can know that “مكتب” reads maktab rather than muktib or mktb. Any tool claiming accurate transliteration is guessing. So the default in Latin mode is to remove Arabic letters and show you which ones went, and the alternative is a rough consonant table that says so, is off by default, and leaves the result editable afterwards.
Does the same title always give the same slug?
Yes. Every step is a fixed rule with no randomness, no time and no dependency on your device settings, so the same title with the same options gives the same result every time on every machine. If you edit the result field by hand, copy and download take what you see rather than what the tool computed.
What is a URL slug?
A slug is the readable part of a URL that identifies a page, usually after the last slash. In https://example.com/blog/arabic-coffee-guide the slug is arabic-coffee-guide. Slugs are usually lower case with words joined by a separator such as -, and a short slug that describes the page is easier to read and share.
How do I generate a slug from an Arabic title?
Type the title and keep “Arabic Unicode” mode: punctuation and symbols are dropped and spaces become the separator. “أفضل قهوة عربية في الرياض 2026” becomes “أفضل-قهوة-عربية-في-الرياض-2026”, 30 characters long, with a 135-character encoded form. If you need Latin letters only, choose “Latin ASCII” mode: “Best Arabic Coffee in Riyadh 2026” becomes best-arabic-coffee-in-riyadh-2026.