Arabic Vocabulary Lists > Arabic Transliteration
Arabic to English transliteration writes Arabic words in Latin letters so you can read them without the script: مَرْحَبًا becomes marhaban. There is no single standard. Texters use Arabizi, with 3 for ع and 7 for ح. US libraries use ALA-LC, German scholars use DIN 31635, and names and menus use plain kh, sh and gh.
Conversion happens in your browser as you type; nothing is sent anywhere. Out of Arabic, add the vowel marks (harakat) if you have them: unvowelled Arabic does not write short vowels, so the result will not have them either. Into Arabic, a single a inside a word is dropped and i, o and u are written as long vowels, which suits most words (7abibi becomes حبيبي) but not all (3indi becomes عيندي, not عندي), so check names and set phrases by eye.
What each mode of the converter does
The converter turns Arabic into plain English letters, Arabizi or ALA-LC-style romanization, and turns Arabizi back into Arabic. Every mode is only as good as the vowel marks you paste. Arabic normally leaves out the short vowels a, i and u, and the tool cannot print what is not there (see why short vowels vanish without harakat).
- Arabic → English letters (the default): kh, sh, gh, th, dh, no digits, no dots. A doubled letter (shadda) is written twice, so مُحَمَّد gives muhammad. ع becomes an apostrophe mid-word (مَعَ ma'a, with) and is dropped at the start, as in English names: عَلِيّ gives ali.
- Arabic → Arabizi: the sample becomes "mar7aban, kayfa 7aluka?" Doubled letters are written once, so مُحَمَّد gives mu7amad.
- Arabizi → Arabic: shukran gives شكراً. Arabizi does not mark vowel length, so this direction guesses. It drops a, e and u inside a word and writes i and o as ي and و: 3indi comes out عيندي, not عندي. Plain English spellings work too, but h and s always become ه and س, never ح or ص. Check the result by eye.
- Arabic → Scholarly: ALA-LC style, "marḥaban, kayfa ḥāluka?" It keeps the grammatical endings you typed, which the Library of Congress rules drop at the end of a sentence (ḥāluk).
Conversion runs in your browser; nothing is uploaded. Swap sends the result back through the Arabizi direction to check a round trip.
Arabic transliteration chart: four systems side by side
Fifteen of the 34 rows below read the same in every column. The other 19 are where the systems part ways. The first two columns are exactly what our converter prints. ALA-LC follows the Library of Congress Arabic romanization table (2012); DIN 31635, the German standard, follows its published letter table. For each letter's four written forms, see the letter shapes and ALA-LC values on the alphabet chart.
| Arabic | Name | Simple English (our default; names, signs, menus) | Arabizi (texting, social media) | ALA-LC (Library of Congress, US libraries) | DIN 31635 (German-language scholarship) |
|---|---|---|---|---|---|
| ا | alif (start of a word) | vowel only | vowel only | vowel only | vowel only |
| ب | ba | b | b | b | b |
| ت | ta | t | t | t | t |
| ث | tha | th | th | th | ṯ |
| ج | jim | j | j | j | ǧ |
| ح | ha (heavy) | h | 7 | ḥ | ḥ |
| خ | kha | kh | kh (also 5, 7') | kh | ḫ |
| د | dal | d | d | d | d |
| ذ | dhal | dh | dh | dh | ḏ |
| ر | ra | r | r | r | r |
| ز | zay | z | z | z | z |
| س | sin | s | s | s | s |
| ش | shin | sh | sh | sh | š |
| ص | sad | s | s (also 9) | ṣ | ṣ |
| ض | dad | d | d (also 9') | ḍ | ḍ |
| ط | ta (heavy) | t | 6 | ṭ | ṭ |
| ظ | za (heavy) | z | z (also 6') | ẓ | ẓ |
| ع | ayn | ' (dropped at the start of a word) | 3 | ʻ | ʿ |
| غ | ghayn | gh | gh (also 3', and 8 in Lebanon) | gh | ġ |
| ف | fa | f | f | f | f |
| ق | qaf | q | q (also 2, g, 8) | q | q |
| ك | kaf | k | k | k | k |
| ل | lam | l | l | l | l |
| م | mim | m | m | m | m |
| ن | nun | n | n | n | n |
| ه | ha | h | h | h | h |
| و | waw (consonant) | w | w | w | w |
| ي | ya (consonant) | y | y | y | y |
| ء | hamza | ' (not written at the start of a word) | 2 | ʼ (not written at the start of a word) | ʾ |
| ة | ta marbuta | a | a | h (t in a possessive phrase) | h (t in a possessive phrase) |
| ى | alif maqsura | a | a | á | ā |
| ـَا | long a | a | a | ā | ā |
| ـُو | long u | u | u | ū | ū |
| ـِي | long i | i | i | ī | ī |
Two details are easy to miss. ALA-LC marks ayn and hamza with comma shapes (ʻ ʼ), DIN with half rings (ʿ ʾ). And ALA-LC always writes the article as al- (the Library of Congress example is al-kitāb al-thānī, the second book), while DIN writes it as spoken before sun letters: aš-šams, the sun.
One sentence in every system
Every system drops something, and one sentence run through all of them shows what. عِنْدِي سُؤَال صَغِير means "I have a small question", and it holds an ayn, a hamza, a heavy s and three long vowels.
| System | Result | What it loses |
|---|---|---|
| Arabic, vowelled | عِنْدِي سُؤَال صَغِير | nothing (the source) |
| Simple English (our default) | indi su'al saghir | the ayn at the start, heavy ṣ against plain s, long vowels against short |
| Arabizi (our tool) | 3indi su2al saghir | heavy ṣ against plain s, vowel length |
| ALA-LC | ʻindī suʼāl ṣaghīr | nothing, but it needs special characters |
| DIN 31635 | ʿindī suʾāl ṣaġīr | nothing; one character per letter (gh is ġ) |
| Simple English from unvowelled Arabic | ndi s'al sghir | every short vowel, and the ayn with it |
Look at the last row. Without vowel marks, every mode of the converter gives you something you cannot pronounce, because the vowels were never on the page. Only someone who already knows the word can put them back.
Arabizi, Franco Arabic and the chat alphabet
Arabizi, Franco Arabic and the Arabic chat alphabet are the same thing: Arabic typed in Latin letters, with digits for the sounds English has no letter for. Each digit looks like the letter it replaces: 3 is ع flipped. The alphabet chart explains why each digit was chosen.
There is no official spelling, and it varies by country. 8 means غ in Lebanese texting but often ق in Saudi texting; our converter reads it as غ. Learners usually meet Arabic slang, which is mostly typed in Arabizi, in this form first. You will need to read it long before you would ever choose to write it.
Which transliteration system should you use?
Use the one your reader expects: simple letters for names, signs and messages to English speakers, Arabizi for reading friends' texts, and ALA-LC, the Library of Congress system, to cite an Arabic book in an English bibliography. DIN 31635 is for German-language scholarship, and unless you read that, you can skip it.
For learning, treat all four as a crutch and drop them early, because every one loses something the script keeps. Our phrases already transliterated in three dialects print the Arabic beside every line. Once you are ready to write the letters yourself, print our Arabic letter writing worksheets, then type Arabic with our on-page keyboard instead of spelling it in Latin.
Frequently asked questions
What does Arabic transliteration mean?
Arabic transliteration means writing Arabic words in Latin letters, sound by sound, so they can be read without knowing the script. It is not translation. Transliterated, مرحبا is marhaban; translated, it is hello.
How do you transliterate Arabic to English?
Pick one system and swap each Arabic letter for its Latin equivalent: kh for خ, sh for ش, gh for غ. Short vowels come from the vowel marks, so paste vowelled text into the converter on this page if you have it. Without the marks, you get consonants and long vowels only.
What is an example of transliteration?
كِتَاب (book) is kitab in simple English letters and kitāb in ALA-LC and DIN 31635, where the macron shows the long a. مُحَمَّد is muhammad in simple letters, mu7amad in Arabizi and muḥammad in ALA-LC.
What is Arabizi or Franco Arabic?
They are two names for the same thing: the chat alphabet Arabic speakers use to text in Latin letters. Digits stand in for sounds English has no letter for: 3 for ع, 7 for ح, 2 for the glottal stop, 6 for ط. There is no fixed spelling, so the same word can be typed several ways.
Why does the tool not add short vowels?
Because ordinary Arabic does not write them. كتب can be kataba (he wrote), kutub (books) or kutiba (it was written), and only the vowel marks or the context tell you which. The converter writes what is on the page, so unvowelled input gives consonants and long vowels, and vowelled input gives the full word.
