Google Neural Machine Translation (GNMT) is a

neural machine translation Neural machine translation (NMT) is an approach to machine translation that uses an artificial neural network to predict the likelihood of a sequence of words, typically modeling entire sentences in a single integrated model. Properties They requi ...

(NMT) system developed by Google and introduced in November 2016, that uses an

artificial neural network Artificial neural networks (ANNs), usually simply called neural networks (NNs) or neural nets, are computing systems inspired by the biological neural networks that constitute animal brains. An ANN is based on a collection of connected unit ...

to increase fluency and accuracy in

Google Translate Google Translate is a multilingual neural machine translation service developed by Google to translate text, documents and websites from one language into another. It offers a website interface, a mobile app for Android and iOS, and an API t ...

. GNMT improves on the quality of translation by applying an example-based (EBMT)

machine translation Machine translation, sometimes referred to by the abbreviation MT (not to be confused with computer-aided translation, machine-aided human translation or interactive translation), is a sub-field of computational linguistics that investigates t ...

method in which the system "learns from millions of examples". GNMT's proposed architecture of system learning was first tested on over a hundred languages supported by Google Translate. With the large end-to-end framework, the system learns over time to create better, more natural translations. GNMT attempts to translate whole sentences at a time, rather than just piece by piece. The GNMT network can undertake

interlingual machine translation Interlingual machine translation is one of the classic approaches to machine translation. In this approach, the source language, i.e. the text to be translated is transformed into an interlingua, i.e., an abstract language-independent representa ...

by encoding the semantics of the sentence, rather than by memorizing phrase-to-phrase translations.

History

The

Google Brain Google Brain is a deep learning artificial intelligence research team under the umbrella of Google AI, a research division at Google dedicated to artificial intelligence. Formed in 2011, Google Brain combines open-ended machine learning research w ...

project was established in 2011 in the "secretive Google X research lab" by Google Fellow

Jeff Dean Jeffrey Adgate "Jeff" Dean (born July 23, 1968) is an American computer scientist and software engineer. Since 2018, he is the lead of Google AI, Google's AI division. Education Dean received a B.S., ''summa cum laude'', from the University o ...

, Google Researcher Greg Corrado, and

Stanford University Stanford University, officially Leland Stanford Junior University, is a private research university in Stanford, California. The campus occupies , among the largest in the United States, and enrolls over 17,000 students. Stanford is consider ...

Computer Science professor

Andrew Ng Andrew Yan-Tak Ng (; born 1976) is a British-born American computer scientist and technology entrepreneur focusing on machine learning and AI. Ng was a co-founder and head of Google Brain and was the former Chief Scientist at Baidu, building ...

. Ng's work has led to some of the biggest breakthroughs at Google and Stanford. In November 2016, Google Neural Machine Translation system (GNMT) was introduced. Since then, Google Translate began using neural machine translation (NMT) in preference to its previous

statistical methods Statistics (from German: ''Statistik'', "description of a state, a country") is the discipline that concerns the collection, organization, analysis, interpretation, and presentation of data. In applying statistics to a scientific, industria ...

(SMT) which had been used since October 2007, with its proprietary, in-house SMT technology. Google Translate's NMT system uses a large artificial neural network capable of

deep learning Deep learning (also known as deep structured learning) is part of a broader family of machine learning methods based on artificial neural networks with representation learning. Learning can be supervised, semi-supervised or unsupervised. De ...

. By using millions of examples, GNMT improves the quality of translation, using broader context to deduce the most relevant translation. The result is then rearranged and adapted to approach grammatically based human language. GNMT's proposed architecture of system learning was first tested on over a hundred languages supported by Google Translate. GNMT did not create its own universal interlingua but rather aimed at finding the commonality between many languages using insights from psychology and linguistics. The new translation engine was first enabled for eight languages: to and from English and French, German, Spanish, Portuguese, Chinese, Japanese, Korean and Turkish in November 2016. In March 2017, three additional languages were enabled: Russian, Hindi and Vietnamese along with Thai for which support was added later. Support for Hebrew and Arabic was also added with help from the Google Translate Community in the same month. In mid April 2017 Google Netherlands announced support for Dutch and other European languages related to English. Further support was added for nine Indian languages: Hindi, Bengali, Marathi, Gujarati, Punjabi, Tamil, Telugu, Malayalam and Kannada at the end of April 2017.

Evaluation

The GNMT system is said to represent an improvement over the former Google Translate in that it will be able handle "zero-shot translation", that is it directly translates one language into another (for example, Japanese to Korean). Google Translate previously first translated the source language into English and then translated the English into the target language rather than translating directly from one language to another. A July 2019 study in ''

Annals of Internal Medicine ''Annals of Internal Medicine'' is an academic medical journal published by the American College of Physicians (ACP). It is one of the most widely cited and influential specialty medical journals in the world. ''Annals'' publishes content relevan ...

'' found that "Google Translate is a viable, accurate tool for translating non–English-language trials". Only one disagreement between reviewers reading machine-translated trials was due to a translation error. Since many medical studies are excluded from systematic reviews because the reviewers do not understand the language, GNMT has the potential to reduce bias and improve accuracy in such reviews.

Languages supported by GNMT

As of December 2021, all of the languages of

support GNMT, with Latin being the most recent addition. #

Afrikaans Afrikaans (, ) is a West Germanic language that evolved in the Dutch Cape Colony from the Dutch vernacular of Holland proper (i.e., the Hollandic dialect) used by Dutch, French, and German settlers and their enslaved people. Afrikaans gra ...

Albanian Albanian may refer to: *Pertaining to Albania in Southeast Europe; in particular: **Albanians, an ethnic group native to the Balkans **Albanian language **Albanian culture **Demographics of Albania, includes other ethnic groups within the country ...

Amharic Amharic ( or ; (Amharic: ), ', ) is an Ethiopian Semitic language, which is a subgrouping within the Semitic branch of the Afroasiatic languages. It is spoken as a first language by the Amharas, and also serves as a lingua franca for all oth ...

Arabic Arabic (, ' ; , ' or ) is a Semitic languages, Semitic language spoken primarily across the Arab world.Semitic languages: an international handbook / edited by Stefan Weninger; in collaboration with Geoffrey Khan, Michael P. Streck, Janet C ...

Armenian Armenian may refer to: * Something of, from, or related to Armenia, a country in the South Caucasus region of Eurasia * Armenians, the national people of Armenia, or people of Armenian descent ** Armenian Diaspora, Armenian communities across the ...

# Azerbaijani #

Basque Basque may refer to: * Basques, an ethnic group of Spain and France * Basque language, their language Places * Basque Country (greater region), the homeland of the Basque people with parts in both Spain and France * Basque Country (autonomous co ...

# Belarusian #

Bengali Bengali or Bengalee, or Bengalese may refer to: *something of, from, or related to Bengal, a large region in South Asia * Bengalis, an ethnic and linguistic group of the region * Bengali language, the language they speak ** Bengali alphabet, the w ...

# Bosnian #

Bulgarian Bulgarian may refer to: * Something of, from, or related to the country of Bulgaria * Bulgarians, a South Slavic ethnic group * Bulgarian language, a Slavic language * Bulgarian alphabet * A citizen of Bulgaria, see Demographics of Bulgaria * Bul ...

# Burmese #

Catalan Catalan may refer to: Catalonia From, or related to Catalonia: * Catalan language, a Romance language * Catalans, an ethnic group formed by the people from, or with origins in, Northern or southern Catalonia Places * 13178 Catalan, asteroid #1 ...

# Cebuano #

Chewa Chewa may refer to: *the Chewa people *the Chewa language Chewa (also known as Nyanja, ) is a Bantu language spoken in much of Southern, Southeast and East Africa, namely the countries of Malawi , where it is an official language, and Mozambiq ...

Chinese Chinese can refer to: * Something related to China * Chinese people, people of Chinese nationality, citizenship, and/or ethnicity **''Zhonghua minzu'', the supra-ethnic concept of the Chinese nation ** List of ethnic groups in China, people of va ...

( Simplified) #

(

Traditional A tradition is a belief or behavior (folk custom) passed down within a group or society with symbolic meaning or special significance with origins in the past. A component of cultural expressions and folklore, common examples include holidays or ...

) # Corsican # Croatian #

Czech Czech may refer to: * Anything from or related to the Czech Republic, a country in Europe ** Czech language ** Czechs, the people of the area ** Czech culture ** Czech cuisine * One of three mythical brothers, Lech, Czech, and Rus' Places *Czech, ...

Danish Danish may refer to: * Something of, from, or related to the country of Denmark People * A national or citizen of Denmark, also called a "Dane," see Demographics of Denmark * Culture of Denmark * Danish people or Danes, people with a Danish a ...

Dutch Dutch commonly refers to: * Something of, from, or related to the Netherlands * Dutch people () * Dutch language () Dutch may also refer to: Places * Dutch, West Virginia, a community in the United States * Pennsylvania Dutch Country People E ...

English English usually refers to: * English language * English people English may also refer to: Peoples, culture, and language * ''English'', an adjective for something of, from, or related to England ** English national ide ...

Esperanto Esperanto ( or ) is the world's most widely spoken constructed international auxiliary language. Created by the Warsaw-based ophthalmologist L. L. Zamenhof in 1887, it was intended to be a universal second language for international communi ...

Estonian Estonian may refer to: * Something of, from, or related to Estonia, a country in the Baltic region in northern Europe * Estonians, people from Estonia, or of Estonian descent * Estonian language * Estonian cuisine * Estonian culture See also

...

Filipino Filipino may refer to: * Something from or related to the Philippines ** Filipino language, standardized variety of 'Tagalog', the national language and one of the official languages of the Philippines. ** Filipinos, people who are citizens of th ...

( Tagalog) #

Finnish Finnish may refer to: * Something or someone from, or related to Finland * Culture of Finland * Finnish people or Finns, the primary ethnic group in Finland * Finnish language, the national language of the Finnish people * Finnish cuisine See also ...

# French # Galician #

Georgian Georgian may refer to: Common meanings * Anything related to, or originating from Georgia (country) ** Georgians, an indigenous Caucasian ethnic group ** Georgian language, a Kartvelian language spoken by Georgians **Georgian scripts, three scrip ...

German German(s) may refer to: * Germany (of or related to) **Germania (historical use) * Germans, citizens of Germany, people of German ancestry, or native speakers of the German language ** For citizens of Germany, see also German nationality law **Ger ...

Greek Greek may refer to: Greece Anything of, from, or related to Greece, a country in Southern Europe: *Greeks, an ethnic group. *Greek language, a branch of the Indo-European language family. **Proto-Greek language, the assumed last common ancestor ...

Gujarati Gujarati may refer to: * something of, from, or related to Gujarat, a state of India * Gujarati people, the major ethnic group of Gujarat * Gujarati language, the Indo-Aryan language spoken by them * Gujarati languages, the Western Indo-Aryan sub- ...

Haitian Creole Haitian Creole (; ht, kreyòl ayisyen, links=no, ; french: créole haïtien, links=no, ), commonly referred to as simply ''Creole'', or ''Kreyòl'' in the Creole language, is a French-based creole language spoken by 10–12million people wor ...

Hausa Hausa may refer to: * Hausa people, an ethnic group of West Africa * Hausa language, spoken in West Africa * Hausa Kingdoms, a historical collection of Hausa city-states * Hausa (horse) or Dongola horse, an African breed of riding horse See also * ...

# Hawaiian #

Hebrew Hebrew (; ; ) is a Northwest Semitic language of the Afroasiatic language family. Historically, it is one of the spoken languages of the Israelites and their longest-surviving descendants, the Jews and Samaritans. It was largely preserved ...

Hindi Hindi (Devanāgarī: or , ), or more precisely Modern Standard Hindi (Devanagari: ), is an Indo-Aryan language spoken chiefly in the Hindi Belt region encompassing parts of northern, central, eastern, and western India. Hindi has been de ...

Hmong Hmong may refer to: * Hmong people, an ethnic group living mainly in Southwest China, Vietnam, Laos, and Thailand * Hmong cuisine * Hmong customs and culture ** Hmong music ** Hmong textile art * Hmong language, a continuum of closely related ton ...

# Hungarian # Icelandic #

Igbo Igbo may refer to: * Igbo people, an ethnic group of Nigeria * Igbo language, their language * anything related to Igboland, a cultural region in Nigeria See also * Ibo (disambiguation) * Igbo mythology * Igbo music * Igbo art * * Igbo-Ukwu, a ...

Indonesian Indonesian is anything of, from, or related to Indonesia, an archipelagic country in Southeast Asia. It may refer to: * Indonesians, citizens of Indonesia ** Native Indonesians, diverse groups of local inhabitants of the archipelago ** Indonesian ...

Irish Irish may refer to: Common meanings * Someone or something of, from, or related to: ** Ireland, an island situated off the north-western coast of continental Europe ***Éire, Irish language name for the isle ** Northern Ireland, a constituent unit ...

Italian Italian(s) may refer to: * Anything of, from, or related to the people of Italy over the centuries ** Italians, an ethnic group or simply a citizen of the Italian Republic or Italian Kingdom ** Italian language, a Romance language *** Regional Ita ...

Japanese Japanese may refer to: * Something from or related to Japan, an island country in East Asia * Japanese language, spoken mainly in Japan * Japanese people, the ethnic group that identifies with Japan through ancestry or culture ** Japanese diaspor ...

# Javanese #

Kannada Kannada (; ಕನ್ನಡ, ), originally romanised Canarese, is a Dravidian language spoken predominantly by the people of Karnataka in southwestern India, with minorities in all neighbouring states. It has around 47 million native s ...

# Kazakh # Khmer #

Kinyarwanda Kinyarwanda, Rwandan or Rwanda, officially known as Ikinyarwanda, is a Bantu language and a dialect of the Rwanda-Rundi language that is spoken in Rwanda and adjacent parts of Burundi, the Democratic Republic of the Congo, Uganda (where there ...

Korean Korean may refer to: People and culture * Koreans, ethnic group originating in the Korean Peninsula * Korean cuisine * Korean culture * Korean language **Korean alphabet, known as Hangul or Chosŏn'gŭl **Korean dialects and the Jeju language ** ...

Kurdish Kurdish may refer to: *Kurds or Kurdish people *Kurdish languages *Kurdish alphabets *Kurdistan, the land of the Kurdish people which includes: **Southern Kurdistan **Eastern Kurdistan **Northern Kurdistan **Western Kurdistan See also * Kurd (dis ...

(

Kurmanji Kurmanji ( ku, کورمانجی, lit=Kurdish, translit=Kurmancî, also termed Northern Kurdish, is the northern dialect of the Kurdish languages, spoken predominantly in southeast Turkey, northwest and northeast Iran, northern Iraq, northern Sy ...

) #

Kyrgyz Kyrgyz, Kirghiz or Kyrgyzstani may refer to: * Someone or something related to Kyrgyzstan *Kyrgyz people *Kyrgyz national games *Kyrgyz language *Kyrgyz culture *Kyrgyz cuisine *Yenisei Kirghiz *The Fuyü Gïrgïs language in Northeastern China ...

# Lao #

Latin Latin (, or , ) is a classical language belonging to the Italic branch of the Indo-European languages. Latin was originally a dialect spoken in the lower Tiber area (then known as Latium) around present-day Rome, but through the power of the ...

# Latvian # Lithuanian #

Luxembourgish Luxembourgish ( ; also ''Luxemburgish'', ''Luxembourgian'', ''Letzebu(e)rgesch''; Luxembourgish: ) is a West Germanic language that is spoken mainly in Luxembourg. About 400,000 people speak Luxembourgish worldwide. As a standard form of th ...

# Macedonian # Malagasy #

Malay Malay may refer to: Languages * Malay language or Bahasa Melayu, a major Austronesian language spoken in Indonesia, Malaysia, Brunei and Singapore ** History of the Malay language, the Malay language from the 4th to the 14th century ** Indonesi ...

Malayalam Malayalam (; , ) is a Dravidian language spoken in the Indian state of Kerala and the union territories of Lakshadweep and Puducherry (Mahé district) by the Malayali people. It is one of 22 scheduled languages of India. Malayalam was des ...

# Maltese # Maori #

Marathi Marathi may refer to: *Marathi people, an Indo-Aryan ethnolinguistic group of Maharashtra, India *Marathi language, the Indo-Aryan language spoken by the Marathi people *Palaiosouda, also known as Marathi, a small island in Greece See also * * ...

# Mongolian # Nepali #

Norwegian Norwegian, Norwayan, or Norsk may refer to: *Something of, from, or related to Norway, a country in northwestern Europe *Norwegians, both a nation and an ethnic group native to Norway *Demographics of Norway *The Norwegian language, including the ...

(

Bokmål Bokmål () (, ; ) is an official written standard for the Norwegian language, alongside Nynorsk. Bokmål is the preferred written standard of Norwegian for 85% to 90% of the population in Norway. Unlike, for instance, the Italian language, there ...

) #

Odia Odia, also spelled Oriya or Odiya, may refer to: * Odia people in Odisha, India * Odia language, an Indian language, belonging to the Indo-Aryan branch of the Indo-European language family * Odia alphabet, a writing system used for the Odia languag ...

Pashto Pashto (,; , ) is an Eastern Iranian language in the Indo-European language family. It is known in historical Persian literature as Afghani (). Spoken as a native language mostly by ethnic Pashtuns, it is one of the two official languages ...

Persian Persian may refer to: * People and things from Iran, historically called ''Persia'' in the English language ** Persians, the majority ethnic group in Iran, not to be conflated with the Iranic peoples ** Persian language, an Iranian language of the ...

Polish Polish may refer to: * Anything from or related to Poland, a country in Europe * Polish language * Poles, people from Poland or of Polish descent * Polish chicken *Polish brothers (Mark Polish and Michael Polish, born 1970), American twin screenwr ...

Portuguese Portuguese may refer to: * anything of, from, or related to the country and nation of Portugal ** Portuguese cuisine, traditional foods ** Portuguese language, a Romance language *** Portuguese dialects, variants of the Portuguese language ** Portu ...

# Punjabi (

Gurmukhi Gurmukhī ( pa, ਗੁਰਮੁਖੀ, , Shahmukhi: ) is an abugida developed from the Laṇḍā scripts, standardized and used by the second Sikh guru, Guru Angad (1504–1552). It is used by Punjabi Sikhs to write the language, commonly r ...

) #

Romanian Romanian may refer to: *anything of, from, or related to the country and nation of Romania **Romanians, an ethnic group **Romanian language, a Romance language ***Romanian dialects, variants of the Romanian language **Romanian cuisine, traditional ...

Russian Russian(s) refers to anything related to Russia, including: *Russians (, ''russkiye''), an ethnic group of the East Slavic peoples, primarily living in Russia and neighboring countries *Rossiyane (), Russian language term for all citizens and peo ...

# Samoan #

Scottish Gaelic Scottish Gaelic ( gd, Gàidhlig ), also known as Scots Gaelic and Gaelic, is a Goidelic language (in the Celtic branch of the Indo-European language family) native to the Gaels of Scotland. As a Goidelic language, Scottish Gaelic, as well as ...

# Serbian #

Shona Shona often refers to: * Shona people, a Southern African people * Shona language, a Bantu language spoken by Shona people today Shona may also refer to: * ''Shona'' (album), 1994 album by New Zealand singer Shona Laing * Shona (given name) * S ...

# Sindhi # Sinhala # Slovak #

Slovenian Slovene or Slovenian may refer to: * Something of, from, or related to Slovenia, a country in Central Europe * Slovene language, a South Slavic language mainly spoken in Slovenia * Slovenes The Slovenes, also known as Slovenians ( sl, Sloven ...

# Somali #

Sotho Sotho may refer to: *Sotho people (or ''Basotho''), an African ethnic group principally resident in South Africa, Lesotho and southern Botswana * Sotho language (''Sesotho'' or ''Southern Sotho''), a Bantu language spoken in southern Africa, an off ...

Spanish Spanish might refer to: * Items from or related to Spain: **Spaniards are a nation and ethnic group indigenous to Spain **Spanish language, spoken in Spain and many Latin American countries **Spanish cuisine Other places * Spanish, Ontario, Cana ...

Sundanese Sundanese may refer to: * Sundanese people * Sundanese language * Sundanese script Standard Sundanese script (''Aksara Sunda Baku'', ) is a writing system which is used by the Sundanese people. It is built based on Old Sundanese script (' ...

# Swahili #

Swedish Swedish or ' may refer to: Anything from or related to Sweden, a country in Northern Europe. Or, specifically: * Swedish language, a North Germanic language spoken primarily in Sweden and Finland ** Swedish alphabet, the official alphabet used by ...

Tajik Tajik, Tadjik, Tadzhik or Tajikistani may refer to: * Someone or something related to Tajikistan * Tajiks, an ethnic group in Tajikistan, Afghanistan and Uzbekistan * Tajik language, the official language of Tajikistan * Tajik (surname) * Tajik cu ...

Tamil Tamil may refer to: * Tamils, an ethnic group native to India and some other parts of Asia **Sri Lankan Tamils, Tamil people native to Sri Lanka also called ilankai tamils **Tamil Malaysians, Tamil people native to Malaysia * Tamil language, nativ ...

Tatar The Tatars ()Tatar
in the Collins English Dictionary is an umbrella term for different

Telugu Telugu may refer to: * Telugu language, a major Dravidian language of India *Telugu people, an ethno-linguistic group of India * Telugu script, used to write the Telugu language ** Telugu (Unicode block), a block of Telugu characters in Unicode S ...

Thai Thai or THAI may refer to: * Of or from Thailand, a country in Southeast Asia ** Thai people, the dominant ethnic group of Thailand ** Thai language, a Tai-Kadai language spoken mainly in and around Thailand *** Thai script *** Thai (Unicode block ...

Turkish Turkish may refer to: *a Turkic language spoken by the Turks * of or about Turkey ** Turkish language *** Turkish alphabet ** Turkish people, a Turkic ethnic group and nation *** Turkish citizen, a citizen of Turkey *** Turkish communities and mi ...

# Turkmen #

Ukrainian Ukrainian may refer to: * Something of, from, or related to Ukraine * Something relating to Ukrainians, an East Slavic people from Eastern Europe * Something relating to demographics of Ukraine in terms of demography and population of Ukraine * So ...

Urdu Urdu (;"Urdu"
''

Uyghur # Uzbek #

Vietnamese Vietnamese may refer to: * Something of, from, or related to Vietnam, a country in Southeast Asia ** A citizen of Vietnam. See Demographics of Vietnam. * Vietnamese people, or Kinh people, a Southeast Asian ethnic group native to Vietnam ** Overse ...

Welsh Welsh may refer to: Related to Wales * Welsh, referring or related to Wales * Welsh language, a Brittonic Celtic language spoken in Wales * Welsh people People * Welsh (surname) * Sometimes used as a synonym for the ancient Britons (Celtic peop ...

# West Frisian #

Xhosa Xhosa may refer to: * Xhosa people, a nation, and ethnic group, who live in south-central and southeasterly region of South Africa * Xhosa language, one of the 11 official languages of South Africa, principally spoken by the Xhosa people See als ...

Yiddish Yiddish (, or , ''yidish'' or ''idish'', , ; , ''Yidish-Taytsh'', ) is a West Germanic language historically spoken by Ashkenazi Jews. It originated during the 9th century in Central Europe, providing the nascent Ashkenazi community with a ver ...

Yoruba The Yoruba people (, , ) are a West African ethnic group that mainly inhabit parts of Nigeria, Benin, and Togo. The areas of these countries primarily inhabited by Yoruba are often collectively referred to as Yorubaland. The Yoruba constitute ...

# Zulu

References

External links

Google’s Neural Machine Translation System: Bridging the Gap between Human and Machine TranslationStatistical Machine TranslationInternational Association for Machine Translation (IAMT)

Machine Translation Archive
by John Hutchins. An electronic repository (and bibliography) of articles, books and papers in the field of machine translation and computer-based translation technology
Machine translation (computer-based translation)
– Publications by John Hutchins (includes

PDF Portable Document Format (PDF), standardized as ISO 32000, is a file format developed by Adobe in 1992 to present documents, including text formatting and images, in a manner independent of application software, hardware, and operating systems. ...

s of several books on machine translation) {{Natural Language Processing Applications of artificial intelligence Computational linguistics Machine translation Artificial neural networks Tasks of natural language processing

Neural Machine Translation Neural machine translation (NMT) is an approach to machine translation that uses an artificial neural network to predict the likelihood of a sequence of words, typically modeling entire sentences in a single integrated model. Properties They requi ...