Arabic Transliteration

Arabic script poses a fundamental challenge for any system that needs to process, index, or compare text alongside Latin-script data. Database lookups, file naming, URL slugs, and linguistic analysis all benefit from having a Latin representation of Arabic text. However, naive transliteration fails on critical details like lam-alef ligatures, which should map to a single unit rather than two separate characters.

This tool uses a Buckwalter-style per-letter mapping that handles the full Arabic character set including the four lam-alef forms, Persian extensions like pe and che, and common tashkeel marks. It is a transliteration system, not a phonetic transcription, meaning each Arabic letter maps predictably to one or two Latin characters without depending on context or dialect pronunciation.

| Arabic Letter | Transliteration | Notes |

| --- | --- | --- |

| b (ba) | b | Single Latin character |

| t (ta) | t | Single Latin character |

| th (tha) | th | Two-character mapping |

| kh (kha) | kh | Two-character mapping |

| ayn (ayn) | ' | Apostrophe representation |

| lam-alef (la) | la | Special-cased ligature |

| lam-alef hamza (la) | lA | Capital A for hamza above |

| pe (Persian) | p | Persian extension |

| che (Persian) | c | Persian extension |

Paste your Arabic text into the right-to-left input box.

Toggle the Capitalize option if you want the Latin output to follow sentence-style capitalization.

Read the transliterated result in the left-to-right output box, where lam-alef ligatures appear as combined units.

Click Copy to grab the transliterated text for use in databases, filenames, or analysis pipelines.

Test with a simple known word: the Arabic word for book should transliterate to 'ktab'. Verify that a lam-alef sequence like the Arabic definite article 'al' transliterates as a combined unit rather than separate 'l' and 'a'. Try a Persian word containing the letter pe to confirm it maps to 'p'. If you enable the Capitalize toggle, check that the first letter of each sentence in the output is uppercase.

Confusing transliteration with phonetic transcription; this tool maps letters, not sounds, so 'a' and 'aa' may look the same to a reader expecting IPA-style output.

Expecting tashkeel (diacritics) to appear in the output; they are stripped before mapping because they have no Latin equivalents.

Assuming the tool handles Arabic presentation forms (initial, medial, final shapes), which are rendering artifacts and not part of the stored text.

Expecting dialect-specific pronunciation differences to be reflected; the mapping is script-level, not phonological.

The four lam-alef ligature forms (lam-alef, lam-alef hamza above, lam-alef hamza below, lam-alef madda) are each special-cased to produce distinct Latin output rather than decomposing into separate lam and alef characters. Persian and Urdu extensions including pe, che, zhe, and gaf are mapped to 'p', 'c', 'zh', and 'g' respectively. Tashkeel marks are stripped entirely before transliteration begins, which is consistent with how most computational Arabic processing pipelines handle vocalized text.

Creating searchable Latin-keyword indexes for Arabic-language content in a multilingual database.

Generating URL slugs from Arabic article titles for a CMS that does not support non-ASCII paths.

Preparing Arabic text for linguistic analysis tools that require Latin-script input.

Building pronunciation guides for Arabic language learners who can read the Latin alphabet.

Q: Which transliteration scheme is used?

A: A Buckwalter-style per-letter map. Each Arabic letter maps to one or two Latin characters (e.g., th for tha, kh for kha, ' for ayn). It is a transliteration, not a phonetic transcription.

Q: Are diacritics (tashkeel) preserved?

A: No. Tashkeel are stripped before mapping because they are combining marks without Latin equivalents. Use the Arabic Tashkeel Remover tool to see bare letters separately.

Q: How are lam-alef ligatures handled?

A: The four lam-alef forms (la, la with hamza above, la with hamza below, la with madda) are special-cased so they transliterate as combined units rather than separate l + a.

Q: Does it support Persian or Urdu letters?

A: Yes. Pe, che, zhe, and gaf are mapped. Other extended Arabic-Persian letters are passed through unchanged.

Q: Is this a phonetic transcription?

A: No. It is a letter-by-letter romanization. For pronunciation you would need a transcription system like ALA-LC or a TTS engine.

Try the Arabic Transliteration tool. See also Arabic Tashkeel Remover, Remove Extra Spaces, and Case Converter.