Compiled by the UNGEGN Working Group on Romanization Systems
Version 4.0, February 2013


The current United Nations recommended system was approved in 2012 (X/8), based on the system called ‘broad transcription’ that had been developed for Persian since 1988, and officially approved in 20031.

Previously the United Nations had approved a romanization system in 1967 (I/13), based on the official system adopted by Iran and published in its English version as Transliteration of Farsi Geographic Names to Latin Alphabet (September 1966)2.

The current system is being implemented in the Islamic Republic of Iran, it is used in the national geographical names database and in some maps.

Persian (Farsi) uses the Perso-Arabic script that is written from right to left. Persian writing usually omits vowel points and diacritical marks, which makes it difficult to obtain uniform results in romanization. The romanization is not fully reversible as it is based on pronunciation and thus different characters pronounced identically are also romanized in the same way. Easier use and better application of the romanization system and compliance with the structure and characteristics of the Persian language were the main reasons for choosing the romanization method.


For the initial, medial and final forms of the principal characters in the romanization table below see the Arabic section of the report.

1 ا، ءA,D,E
2 ب b
3 پ p
4 ت t
5 ث s
6 ج j
7 چ č
8 ح h
9 خ x
10 د d
11 ذ z
12 ر r
13 ز z
14 ژ ž
15 س s
16 ش š
17 ص s
18 ض z
19 ط t
20 ظ z
21 ع B
22 غ q
23 ف f
24 ق q
25 ک k
26 گ g
27 ل l
28 م m
29 ن n
30 و vD
31 ه hC
32 ی yD

Vowel characters and diacritical marks (ب stands for any consonant)

1 بَ a
2 بِ e
3 بُ o
4 بٰا ā
5 بٰى ā
6 آ ā
7 بى i
8 بو u
9 بِى ey
10 بُو ow
11 بْ (A)
12 بّ (B)
13 ـ، بى -eC, -yeD


  1. The adjectival ending ـيه in Persian is romanized -iye. The definite article ال is romanized ol. When the definite article precedes a word beginning with d, n, r, s, š, t, z or ž, the l of the definite article is assimilated with the first consonant of the word: e.g. زين الدين Zeyn od Din.
  2. The relational suffix (ezāfe) is romanized -e after a final consonant character (except for ه, when silent) e.g. گل سرخ Gol-e Sorx. It is usually not expressed in Persian writing after a consonant. After vowel characters and silent ه, it is written with ى and romanized -ye, e.g. پاى آب Pā-ye Āb, دهنه‌ى ممبر (may be written دهنهٴ ممبر) Dahane-ye Mambar.

Other systems of romanization

The previous Iranian national system (1966), approved by the United Nations in 1967 and coinciding with that of BGN/PCGN 1958 (revised 2007) was mainly a transliteration system whereby several Persian characters were distinguished and in some cases digraphs were used:

ā (ى)= á
č= ch
-e (ه)= -eh
h (ح)= ḩ
i= ī
q (غ)= gh
s (ص)= ş
s (ث)= s̄
š= sh
t (ط)= ţ
u= ū
x= kh
z (ظ)= z̧
z (ذ)= z̄
z (ض)= ẕ
ž= zh
’ (ع)= ‘

The simplified transliteration of Persian in the standard ISO 233-3:1999 has the following equivalents to the UN system:

ā = â
h (ح)= ḥ
i = y?
q (غ)= ġ
s (ث)= s̱
s (ص)= ṣ
t (ط)= ṭ
u = v?
x= ḵ
z (ذ)= ẕ
z (ض)= ż
z (ظ)= ẓ*
’ (ع)= ‘

*Note. The ISO standard obviously contains some typographic errors which are corrected and indicated by an asterisk in the table above. Consequently, not all the correspondences here can be considered as verified.


  1. New Persian romanization system. Tenth United Nations Conference on the Standardization of Geographical Names. New York, 31 July – 9 August 2012. Document E/CONF.101/118/Rev.1.
  2. Toponymic Guidelines for map and other editors – Revised edition 1998. Submitted by the Islamic Republic of Iran. UNGEGN, 20th session. New York, 17–28 January 2000, Working Paper No. 41.