Language support

Neural Machine Translation model

The Translation API's recognition engine supports a wide variety of languages for the Neural Machine Translation (NMT) model.

These languages are specified within a recognition request using language code parameters as noted on this page. Most language code parameters conform to ISO-639 identifiers, except where noted.

Translations from any language to any language in this list are supported. Romanization and transliteration are supported only on the Cloud Translation - Advanced API. The list is updated as new languages are added. You can also list the supported languages by using Cloud Translation - Basic or Cloud Translation - Advanced APIs.

Try it for yourself

If you're new to Google Cloud, create an account to evaluate how Cloud Translation performs in real-world scenarios. New customers also get $300 in free credits to run, test, and deploy workloads.

Try Cloud Translation free

Language ISO-639 code
Abkhaz ab
Acehnese ace
Acholi ach
Afrikaans af
Albanian sq
Alur alz
Amharic am
Arabic ar
Armenian hy
Assamese as
Awadhi awa
Aymara ay
Azerbaijani az
Balinese ban
Bambara bm
Bashkir ba
Basque eu
Batak Karo btx
Batak Simalungun bts
Batak Toba bbc
Belarusian be
Bemba bem
Bengali bn
Betawi bew
Bhojpuri bho
Bikol bik
Bosnian bs
Breton br
Bulgarian bg
Buryat bua
Cantonese yue
Catalan ca
Cebuano ceb
Chichewa (Nyanja) ny
Chinese (Simplified) zh-CN or zh (BCP-47)
Chinese (Traditional) zh-TW (BCP-47)
Chuvash cv
Corsican co
Crimean Tatar crh
Croatian hr
Czech cs
Danish da
Dinka din
Divehi dv
Dogri doi
Dombe dov
Dutch nl
Dzongkha dz
English en
Esperanto eo
Estonian et
Ewe ee
Fijian fj
Filipino (Tagalog) fil or tl
Finnish fi
French fr
French (French) fr-FR
French (Canadian) fr-CA
Frisian fy
Fulfulde ff
Ga gaa
Galician gl
Ganda (Luganda) lg
Georgian ka
German de
Greek el
Guarani gn
Gujarati gu
Haitian Creole ht
Hakha Chin cnh
Hausa ha
Hawaiian haw
Hebrew iw or he
Hiligaynon hil
Hindi hi
Hmong hmn
Hungarian hu
Hunsrik hrx
Icelandic is
Igbo ig
Iloko ilo
Indonesian id
Irish ga
Italian it
Japanese ja
Javanese jw or jv
Kannada kn
Kapampangan pam
Kazakh kk
Khmer km
Kiga cgg
Kinyarwanda rw
Kituba ktu
Konkani gom
Korean ko
Krio kri
Kurdish (Kurmanji) ku
Kurdish (Sorani) ckb
Kyrgyz ky
Lao lo
Latgalian ltg
Latin la
Latvian lv
Ligurian lij
Limburgan li
Lingala ln
Lithuanian lt
Lombard lmo
Luo luo
Luxembourgish lb
Macedonian mk
Maithili mai
Makassar mak
Malagasy mg
Malay ms
Malay (Jawi) ms-Arab
Malayalam ml
Maltese mt
Maori mi
Marathi mr
Meadow Mari chm
Meiteilon (Manipuri) mni-Mtei
Minang min
Mizo lus
Mongolian mn
Myanmar (Burmese) my
Ndebele (South) nr
Nepalbhasa (Newari) new
Nepali ne
Northern Sotho (Sepedi) nso
Norwegian no
Nuer nus
Occitan oc
Odia (Oriya) or
Oromo om
Pangasinan pag
Papiamento pap
Pashto ps
Persian fa
Polish pl
Portuguese pt
Portuguese (Portugal) pt-PT
Portuguese (Brazil) pt-BR
Punjabi pa
Punjabi (Shahmukhi) pa-Arab
Quechua qu
Romani rom
Romanian ro
Rundi rn
Russian ru
Samoan sm
Sango sg
Sanskrit sa
Scots Gaelic gd
Serbian sr
Sesotho st
Seychellois Creole crs
Shan shn
Shona sn
Sicilian scn
Silesian szl
Sindhi sd
Sinhala (Sinhalese) si
Slovak sk
Slovenian sl
Somali so
Spanish es
Sundanese su
Swahili sw
Swati ss
Swedish sv
Tajik tg
Tamil ta
Tatar tt
Telugu te
Tetum tet
Thai th
Tigrinya ti
Tsonga ts
Tswana tn
Turkish tr
Turkmen tk
Twi (Akan) ak
Ukrainian uk
Urdu ur
Uyghur ug
Uzbek uz
Vietnamese vi
Welsh cy
Xhosa xh
Yiddish yi
Yoruba yo
Yucatec Maya yua
Zulu zu

Romanization and transliteration support

Romanization and transliteration support the following languages.

Language ISO-639 code Romanization Transliteration
Arabic ar
Amharic am
Bengali bn
Belarusian be
Gujarati gu
Hindi hi
Japanese ja
Kannada kn
Myanmar uk
Russian ru
Serbian sr
Tamil ta
Telugu te
Ukrainian uk

Custom models

For Cloud Translation - Advanced, AutoML Translation supports the following language pairs for custom models. A supported language means that Google has an existing neural machine translation (NMT) model for that language, which AutoML Translation uses as a base to create a custom model during training.

Language Pair Language Codes
Afrikaans <-> English af <-> en
Albanian <-> English sq <-> en
Arabic <-> English ar <-> en
Azerbaijani <-> English az <-> en
Bengali <-> English bn <-> en
Bulgarian <-> English bg <-> en
Catalan <-> English ca <-> en
Chinese (Simplified) <-> English zh-CN * <-> en
Chinese (Traditional) <-> English zh-TW <-> en
Croatian <-> English hr <-> en
Czech <-> English cs <-> en
Danish <-> English da <-> en
Dutch <-> English nl <-> en
Estonian <-> English et <-> en
Finnish <-> English fi <-> en
French <-> English fr <-> en
Galician <-> English gl <-> en
Georgian <-> English ka <-> en
German <-> English de <-> en
Greek <-> English el <-> en
Gujarati <-> English gu <-> en
Haitian Creole <-> English ht <-> en
Hebrew <-> English iw <-> en
Hindi <-> English hi <-> en
Hungarian <-> English hu <-> en
Icelandic <-> English is <-> en
Indonesian <-> English id <-> en
Italian <-> English it <-> en
Japanese <-> English ja <-> en
Korean <-> English ko <-> en
Latvian <-> English lv <-> en
Lithuanian <-> English lt <-> en
Malay <-> English ms <-> en
Marathi <-> English mr <-> en
Norwegian <-> English no <-> en
Persian <-> English fa <-> en
Polish <-> English pl <-> en
Portuguese <-> English pt <-> en
Punjabi <-> English pa <-> en
Romanian <-> English ro <-> en
Russian <-> English ru <-> en
Serbian <-> English sr <-> en
Slovak <-> English sk <-> en
Slovenian <-> English sl <-> en
Spanish <-> English es <-> en
Swahili <-> English sw <-> en
Swedish <-> English sv <-> en
Thai <-> English th <-> en
Turkish <-> English tr <-> en
Ukrainian <-> English uk <-> en
Urdu <-> English ur <-> en
Vietnamese <-> English vi <-> en
Welsh <-> English cy <-> en

* Simplified Chinese can be specified either by zh-CN or zh.

Supported codes for language variants

The following language codes are variants of the supported languages in the previous table. You can use these codes as the source or target language when you create datasets.

Google doesn't have base NMT models for these languages. Instead, Cloud Translation uses the language variant's associated base model for training custom models.

These codes are useful, for example, when you translate content for a particular dialect or region. For example, suppose you have localized data for zh-HK that you create a custom model for. When you perform translations, you can specify the zh-HK language code, which points to your custom model and produces more accurate translations for that locale.

The following table lists the language codes, their descriptions, and their associated base models that Cloud Translation uses when training custom models.

Language code Description Base model
zh-HK Hong Kong (Traditional) zh-TW

Translation LLM supported languages

The following languages pairs are supported for text translations that use the translation LLM and for adaptive translations.

Language pair Language codes
Arabic <-> English ar <-> en
Chinese (Simplified) <-> English zh-CN <-> en
Czech <-> English cs <-> en
Dutch <-> English nl <-> en
French <-> English fr <-> en
German <-> English de <-> en
Hindi <-> English hi <-> en
Indonesian <-> English id <-> en
Italian <-> English it <-> en
Japanese <-> English ja <-> en
Korean <-> English ko <-> en
Polish <-> English pl <-> en
Portuguese <-> English pt <-> en
Russian <-> English ru <-> en
Spanish <-> English es <-> en
Thai <-> English th <-> en
Turkish <-> English tr <-> en
Ukrainian <-> English uk <-> en
Vietnamese <-> English vi <-> en