View source for Machine translation

{{short description|Computerized translation between natural languages}}
{{distinguish|Computer-assisted translation|Interactive machine translation|Translator (computing)}}
{{Use dmy dates|date=July 2014}}
[[File:WordLensDemo5Feb2012.jpg|thumb|upright=1.3|alt=A person is holding a sign reads "BIENVENIDO AL FUTURO". Another person is holding an iPhone to translate the sign. The phone shows "WELCOME TO THE FUTURE", and "Spanish to English".|A mobile phone app translating Spanish text into English]]
{{Translation sidebar}}
'''Machine translation''' is the use of computational techniques to [[translation|translate]] text or speech from one [[language]] to another, including the contextual, idiomatic, and pragmatic nuances of both languages.

Machine translation tools, while some language models are capable of generating comprehensible results, remain limited by the complexity of language and emotion, often lacking depth and semantic precision. Its quality is influenced by linguistic, grammatical, tonal, and cultural differences, making it inadequate to replace real translators fully.<ref>{{Cite web |title=Can Machine Translation Replace Human Translation In The Future? |url=https://www.linkedin.com/pulse/can-machine-translation-replace-human-future-gte-localize-6petc |access-date=2025-10-09 |website=www.linkedin.com |language=en}}</ref><ref>{{Cite book |last=Mohammed H. Al Aqad |title=Role of AI in Translation and Interpretation |date=2025-04-25 |publisher=IGI global |isbn=9798337300603 |pages=454}}</ref> Effective improvement requires understanding the target society’s customs and historical context, human intervention and visual cues remain necessary in [[simultaneous interpretation]], on the other hand, domain-specific customization, such as for technical documentation or official texts—can yield more stable results<ref>{{Cite web |last=Anichini |first=Nicole |date=2024-01-25 |title=The Pros and Cons of Machine Translation |url=https://mastertcloc.unistra.fr/2024/01/25/the-pros-and-cons-of-machine-translation/ |access-date=2025-10-09 |website=TCLoc Master's |language=en-US}}</ref>, and is commonly employed in multilingual websites and professional databases.<ref>{{Cite web |last=Magdalena Cieślak |title=THE SCOPE AND LIMITS OF MACHINE TRANSLATION |url=https://ojs.letras.up.pt/index.php/elingUP/article/download/2528/2316/6148 |archive-url=http://web.archive.org/web/20210627185058/https://ojs.letras.up.pt/index.php/elingUP/article/download/2528/2316/6148 |archive-date=2021-06-27 |access-date=2025-10-09 |website=ojs.letras.up.pt}}</ref><ref>{{Cite journal |last=Naveen |first=Palanichamy |last2=Trojovský |first2=Pavel |date=2024-10-18 |title=Overview and challenges of machine translation for contextually appropriate translations |url=https://www.sciencedirect.com/science/article/pii/S2589004224021035 |journal=iScience |volume=27 |issue=10 |article-number=110878 |doi=10.1016/j.isci.2024.110878 |issn=2589-0042|pmc=11465115 }}</ref>

Early approaches were mostly [[Rule-based machine translation|rule-based]] or [[Statistical machine translation|statistical]]. These methods have since been superseded by [[neural machine translation]]<ref>{{Cite web |date=3 October 2016 |title=Google Translate Gets a Deep-Learning Upgrade |url=https://spectrum.ieee.org/google-translate-gets-a-deep-learning-upgrade |access-date=2024-07-07 |website=IEEE Spectrum |language=en}}</ref> and [[large language model]]s.<ref>{{Cite web |date=2024-02-23 |title=Google Translate vs. ChatGPT: Which One Is the Best Language Translator? |url=https://uk.pcmag.com/ai/151950/google-translate-vs-chatgpt-which-one-is-the-best-language-translator |access-date=2024-07-07 |website=PCMag UK |language=en-gb}}</ref>

==History==
{{Main|History of machine translation}}

===Origins===
The origins of machine translation can be traced back to the work of [[Al-Kindi]], a ninth-century Arabic [[cryptographer]] who developed techniques for systemic language translation, including [[cryptanalysis]], [[frequency analysis]], and [[probability]] and [[statistics]], which are used in modern machine translation.<ref>{{Cite web |last=DuPont |first=Quinn |date=January 2018 |title=The Cryptological Origins of Machine Translation: From al-Kindi to Weaver |url=http://amodern.net/article/cryptological-origins-machine-translation/ |archive-url=https://web.archive.org/web/20190814061915/http://amodern.net/article/cryptological-origins-machine-translation/ |archive-date=14 August 2019 |access-date=2 September 2019 |website=Amodern}}</ref> The idea of machine translation later appeared in the 17th century. In 1629, [[René Descartes]] proposed a universal language, with equivalent ideas in different tongues sharing one symbol.<ref>{{Cite book |last=Knowlson |first=James |title=Universal Language Schemes in England and France, 1600-1800 |date=1975 |publisher=University of Toronto Press |isbn=0-8020-5296-7 |location=Toronto}}</ref>

The idea of using digital computers for translation of natural languages was proposed as early as 1947 by England's [[Andrew Donald Booth|A. D. Booth]]<ref>{{Cite book |last=Booth |first=Andrew D. |url=https://archive.org/details/sim_computers-and-people_1953-05_2_4/page/n8/ |title=Computers and Automation 1953-05: Vol 2 Iss 4 |date=1953-05-01 |publisher=Berkeley Enterprises |page=6 |language=en |chapter=MECHANICAL TRANSLATION}}</ref> and [[Warren Weaver]] at [[Rockefeller Foundation]] in the same year. "The memorandum written by [[Warren Weaver]] in 1949 is perhaps the single most influential publication in the earliest days of machine translation."<ref>{{cite book
   |url=https://pdfs.semanticscholar.org/eaa9/ccf94b4d129c26faf45a1353ffcbbe9d4fda.pdf
   |archive-url=https://web.archive.org/web/20200228015454/https://pdfs.semanticscholar.org/eaa9/ccf94b4d129c26faf45a1353ffcbbe9d4fda.pdf
   |archive-date=2020-02-28
   |chapter=Warren Weaver and the launching of MT |via=[[Semantic Scholar]]  |author=J. Hutchins|title=Early Years in Machine Translation |series=Studies in the History of the Language Sciences |year=2000 |volume=97 |page=17 |doi=10.1075/sihols.97.05hut |isbn=978-90-272-4586-1 |s2cid=163460375 }}</ref><ref>{{cite web
|url=https://www.britannica.com/biography/Warren-Weaver
|title=Warren Weaver, American mathematician
|date=July 13, 2020
|access-date=7 August 2020
|archive-date=6 March 2021
|archive-url=https://web.archive.org/web/20210306061225/https://www.britannica.com/biography/Warren-Weaver
|url-status=live
}}</ref> Others followed. A demonstration was made in 1954 on the [[APEXC]] machine at [[Birkbeck, University of London|Birkbeck College]] ([[University of London]]) of a rudimentary translation of English into French. Several papers on the topic were published at the time, and even articles in popular journals (for example an article by Cleave and Zacharov in the September 1955 issue of ''[[Wireless World]]'').  A similar application, also pioneered at Birkbeck College at the time, was reading and composing [[Braille]] texts by computer.

===1950s===
The first researcher in the field, [[Yehoshua Bar-Hillel]], began his research at MIT (1951). A [[Georgetown University]] MT research team, led by Professor Michael Zarechnak, followed (1951) with a public demonstration of its [[Georgetown-IBM experiment]] system in 1954. MT research programs popped up in Japan<ref>{{cite book|last1=上野|first1=俊夫|title=パーソナルコンピュータによる機械翻訳プログラムの制作|date=1986-08-13|publisher=(株)ラッセル社|location=Tokyo|isbn=4-947627-00-X|page=16|language=ja|quote=わが国では1956年、当時の電気試験所が英和翻訳専用機「ヤマト」を実験している。この機械は1962年頃には中学1年の教科書で90点以上の能力に達したと報告されている。(translation (assisted by [[Google Translate]]): In 1959 Japan, the [[w:jp:電気試験所|National Institute of Advanced Industrial Science and Technology]](AIST) tested the proper English-Japanese translation machine ''Yamato'', which reported in 1964 as that reached the power level over the score of 90-point on the textbook of first grade of junior hi-school.)}}</ref><ref>{{Cite web | url=http://museum.ipsj.or.jp/computer/dawn/0027.html | title=機械翻訳専用機「やまと」-コンピュータ博物館 | access-date=4 April 2017 | archive-date=19 October 2016 | archive-url=https://web.archive.org/web/20161019171540/http://museum.ipsj.or.jp/computer/dawn/0027.html | url-status=live }}</ref> and Russia (1955), and the first MT conference was held in London (1956).<ref name="Nye">{{cite journal|last1=Nye|first1=Mary Jo|title=Speaking in Tongues: Science's centuries-long hunt for a common language|journal=Distillations|date=2016|volume=2|issue=1|pages=40–43|url=https://www.sciencehistory.org/distillations/magazine/speaking-in-tongues|access-date=20 March 2018|archive-date=3 August 2020|archive-url=https://web.archive.org/web/20200803130801/https://www.sciencehistory.org/distillations/magazine/speaking-in-tongues|url-status=live}}</ref><ref name="Babel">{{cite book|last1=Gordin|first1=Michael D.|title=Scientific Babel: How Science Was Done Before and After Global English|date=2015|publisher=University of Chicago Press|location=Chicago, Illinois|isbn=978-0-226-00029-9}}</ref>

[[David G. Hays]] "wrote about computer-assisted language processing as early as 1957" and "was project leader on computational linguistics
at [[RAND Corporation|Rand]] from 1955 to 1968."<ref>{{cite news
|newspaper=[[The New York Times]]
|quote=wrote about computer-assisted language processing as early as 1957.. was project leader on computational linguistics at Rand from 1955 to 1968.
|url=https://www.nytimes.com/1995/07/28/obituaries/david-g-hays-66-a-developer-of-language-study-by-computer.html
|title=David G. Hays, 66, a Developer Of Language Study by Computer
|author=Wolfgang Saxon
|date=July 28, 1995
|access-date=7 August 2020
|archive-date=7 February 2020
|archive-url=https://web.archive.org/web/20200207035914/https://www.nytimes.com/1995/07/28/obituaries/david-g-hays-66-a-developer-of-language-study-by-computer.html
|url-status=live
}}</ref>

===1960–1975===
Researchers continued to join the field as the Association for Machine Translation and Computational Linguistics was formed in the U.S. (1962) and the National Academy of Sciences formed the Automatic Language Processing Advisory Committee (ALPAC) to study MT (1964). Real progress was much slower, however, and after the [[ALPAC|ALPAC report]] (1966), which found that the ten-year-long research had failed to fulfill expectations, funding was greatly reduced.<ref name="ueno">{{cite book
|last1=上野 |first1=俊夫 |title=パーソナルコンピュータによる機械翻訳プログラムの制作 |date=1986-08-13 |publisher=(株)ラッセル社|isbn=4-947627-00-X|page=16|location=Tokyo|language=ja}}</ref> According to a 1972 report by the Director of Defense Research and Engineering (DDR&E), the feasibility of large-scale MT was reestablished by the success of the Logos MT system in translating military manuals into Vietnamese during that conflict.

The French Textile Institute also used MT to translate abstracts from and into French, English, German and Spanish (1970); Brigham Young University started a project to translate Mormon texts by automated translation (1971).

===1975-1980s ===
[[SYSTRAN]], which "pioneered the field under contracts from the U.S. government"<ref name="MT1998.EmptyAtlantic">{{Cite magazine |last=Budiansky |first=Stephen  |date=December 1998 |title=Lost in Translation |magazine=[[Atlantic Magazine]] |pages=81–84}}</ref> in the 1960s, was used by Xerox to translate technical manuals (1978). Beginning in the late 1980s, as [[computation]]al power increased and became less expensive, more interest was shown in [[statistical machine translation|statistical models for machine translation]]. MT became more popular after the advent of computers.<ref>{{Cite book|title=Conceptual Information Processing|last=Schank|first=Roger C.|date=2014|publisher=Elsevier|isbn=978-1-4832-5879-9|location=New York|page=5}}</ref> SYSTRAN's first implementation system was implemented in 1988 by the online service of the [[La Poste (France)|French Postal Service]] called Minitel.<ref>{{Cite book|title=Machine Translation and the Information Soup: Third Conference of the Association for Machine Translation in the Americas, AMTA'98, Langhorne, PA, USA, October 28–31, 1998 Proceedings|last1=Farwell|first1=David|last2=Gerber|first2=Laurie|last3=Hovy|first3=Eduard|date=2003-06-29|publisher=Springer|isbn=3-540-65259-0|location=Berlin|page=276}}</ref> Various computer based translation companies were also launched, including Trados (1984), which was the first to develop and market Translation Memory technology (1989), though this is not the same as MT. The first commercial MT system for Russian / English / German-Ukrainian was developed at Kharkov State University (1991).

=== 1990s and early 2000s ===
By 1998, "for as little as $29.95" one could "buy a program for translating in one direction between English and a major European language of
your choice" to run on a PC.<ref name=MT1998.EmptyAtlantic/>

MT on the web started with SYSTRAN offering free translation of small texts (1996) and then providing this via [[Babel Fish (website)|AltaVista Babelfish]],<ref name=MT1998.EmptyAtlantic/> which racked up 500,000 requests a day (1997).<ref>{{Cite web |url=https://digital.com/about/babel-fish/ |title=Babel Fish: What Happened To The Original Translation Application?: We Investigate |last1=Barron |first1=Brenda |date=November 18, 2019 |website=Digital.com |language=en-US |access-date=2019-11-22 |archive-date=20 November 2019 |archive-url=https://web.archive.org/web/20191120032732/https://digital.com/about/babel-fish/ |url-status=live }}</ref> The second free translation service on the web was [[Lernout & Hauspie]]'s GlobaLink.<ref name=MT1998.EmptyAtlantic/> ''Atlantic Magazine'' wrote in 1998 that "Systran's Babelfish and GlobaLink's Comprende" handled
"Don't bank on it" with a "competent performance."<ref>and gave other examples too</ref>

[[Franz Josef Och]] (the future head of Translation Development AT Google) won DARPA's speed MT competition (2003).<ref>{{Cite book
|title=Routledge Encyclopedia of Translation Technology |last=Chan |first=Sin-Wai
|date=2015 |publisher=Routledge |isbn=978-0-415-52484-1 |location=Oxon |page=385}}</ref> More innovations during this time included MOSES, the open-source statistical MT engine (2007), a text/SMS translation service for mobiles in Japan (2008), and a mobile phone with built-in speech-to-speech translation functionality for English, Japanese and Chinese (2009). In 2012, Google announced that [[Google Translate]] translates roughly enough text to fill 1 million books in one day.

=== ANNs & LLMs in 2020s ===
{{Empty section|date=November 2025}}

==Approaches==
{{See also|Hybrid machine translation|Example-based machine translation|}}

Before the advent of [[deep learning]] methods, statistical methods required a lot of rules accompanied by [[morphology (linguistics)|morphological]], [[syntax|syntactic]], and [[semantics|semantic]] annotations.

===Rule-based===
{{Main|Rule-based machine translation}}

The rule-based machine translation approach was used mostly in the creation of [[dictionaries]] and grammar programs. Its biggest downfall was that everything had to be made explicit: orthographical variation and erroneous input must be made part of the source language analyser in order to cope with it, and lexical selection rules must be written for all instances of ambiguity.

====Transfer-based machine translation====
{{Main|Transfer-based machine translation}}

Transfer-based machine translation was similar to [[interlingual machine translation]] in that it created a translation from an intermediate representation that simulated the meaning of the original sentence. Unlike interlingual MT, it depended partially on the language pair involved in the translation.

====Interlingual====
{{Main|Interlingual machine translation}}

Interlingual machine translation was one instance of rule-based machine-translation approaches.  In this approach, the source language, i.e. the text to be translated, was transformed into an interlingual language, i.e. a "language neutral" representation that is independent of any language. The target language was then generated out of the [[interlinguistics|interlingua]]. The only interlingual machine translation system that was made operational at the commercial level was the KANT system (Nyberg and Mitamura, 1992), which was designed to translate Caterpillar Technical English (CTE) into other languages.

====Dictionary-based====
{{Main|Dictionary-based machine translation}}

Machine translation used a method based on [[dictionary]] entries, which means that the words were translated as they are by a dictionary.

===Statistical===
{{main|Statistical machine translation}}
Statistical machine translation tried to generate translations using [[statistical methods]] based on bilingual text corpora, such as the [[Hansard#Translation|Canadian Hansard]] corpus, the English-French record of the Canadian parliament and [[Europarl corpus|EUROPARL]], the record of the [[European Parliament]]. Where such corpora were available, good results were achieved translating similar texts, but such corpora were rare for many language pairs. The first statistical machine translation software was [[CANDIDE]] from [[IBM]]. In 2005, Google improved its internal translation capabilities by using approximately 200 billion words from United Nations materials to train their system; translation accuracy improved.<ref>{{cite web |url=http://blog.outer-court.com/archive/2005-05-22-n83.html |title=Google Translator: The Universal Language |publisher=Blog.outer-court.com |date=25 January 2007 |access-date=2012-06-12 |archive-date=20 November 2008 |archive-url=https://web.archive.org/web/20081120030225/http://blog.outer-court.com/archive/2005-05-22-n83.html |url-status=live }}</ref>

SMT's biggest downfall included it being dependent upon huge amounts of parallel texts, its problems with morphology-rich languages (especially with translating ''into'' such languages), and its inability to correct singleton errors.

Some work has been done in the utilization of multiparallel [[text corpus|corpora]], that is a body of text that has been translated into 3 or more languages. Using these methods, a text that has been translated into 2 or more languages may be utilized in combination to provide a more accurate translation into a third language compared with if just one of those source languages were used alone.<ref>{{Cite conference |last=Schwartz |first=Lane |date=2008 |title=Multi-Source Translation Methods |url=https://dowobeha.github.io/papers/amta08.pdf |conference=Paper presented at the 8th Biennial Conference of the Association for Machine Translation in the Americas |archive-url=https://web.archive.org/web/20160629171944/http://dowobeha.github.io/papers/amta08.pdf |archive-date=29 June 2016 |access-date=3 November 2017 |url-status=live}}</ref><ref>{{Cite conference |last1=Cohn |first1=Trevor |last2=Lapata |first2=Mirella |date=2007 |title=Machine Translation by Triangulation: Making Effective Use of Multi-Parallel Corpora |url=http://homepages.inf.ed.ac.uk/mlap/Papers/acl07.pdf |conference=Paper presented at the 45th Annual Meeting of the Association for Computational Linguistics, June 23–30, 2007, Prague, Czech Republic |archive-url=https://web.archive.org/web/20151010171334/http://homepages.inf.ed.ac.uk/mlap/Papers/acl07.pdf |archive-date=10 October 2015 |access-date=3 February 2015 |url-status=live}}</ref><ref>{{Cite journal |last1=Nakov |first1=Preslav |last2=Ng |first2=Hwee Tou |date=2012 |title=Improving Statistical Machine Translation for a Resource-Poor Language Using Related Resource-Rich Languages |url=https://jair.org/index.php/jair/article/view/10764 |journal=Journal of Artificial Intelligence Research |volume=44 |pages=179–222 |arxiv=1401.6876 |doi=10.1613/jair.3540 |doi-access=free}}</ref>

=== Neural MT ===
{{Main|Neural machine translation}}

A [[deep learning]]-based approach to MT, [[neural machine translation]] has made rapid progress in recent years. However, the current consensus is that the so-called human parity achieved is not real, being based wholly on limited domains, language pairs, and certain test benchmarks<ref>Antonio Toral, Sheila Castilho, Ke Hu, and Andy
Way. 2018. Attaining the unattainable? reassessing claims of human parity in neural machine translation. CoRR, abs/1808.10432.</ref> i.e., it lacks statistical significance power.<ref>{{Cite arXiv |eprint=1906.09833 |first1=Graham |last1=Yvette |first2=Haddow |last2=Barry |title=Translationese in Machine Translation Evaluation |date=2019 |last3=Koehn |first3=Philipp|class=cs.CL }}</ref>

Translations by neural MT tools like [[DeepL Translator]], which is thought to usually deliver the best machine translation results as of 2022, typically still need post-editing by a human.<ref>{{cite journal |last1=Katsnelson |first1=Alla |title=Poor English skills? New AIs help researchers to write better |journal=Nature |pages=208–209 |language=en |doi=10.1038/d41586-022-02767-9 |date=29 August 2022|volume=609 |issue=7925 |pmid=36038730 |bibcode=2022Natur.609..208K |s2cid=251931306 |doi-access=free }}</ref><ref>{{cite web |last1=Korab |first1=Petr |title=DeepL: An Exceptionally Magnificent Language Translator |url=https://towardsdatascience.com/deepl-an-exceptionally-magnificent-language-translator-78e86d8062d3 |website=Medium |access-date=9 January 2023 |language=en |date=18 February 2022}}</ref><ref>{{cite news |title=DeepL outperforms Google Translate – DW – 12/05/2018 |url=https://www.dw.com/en/deepl-cologne-based-startup-outperforms-google-translate/a-46581948 |access-date=9 January 2023 |work=Deutsche Welle |language=en}}</ref>

Instead of training specialized translation models on parallel datasets, one can also [[Prompt engineering|directly prompt]] generative [[large language model]]s like [[Generative pre-trained transformer|GPT]] to translate a text.<ref name="Hendy2023">{{cite arXiv |last1=Hendy |first1=Amr |last2=Abdelrehim |first2=Mohamed |last3=Sharaf |first3=Amr |last4=Raunak |first4=Vikas |last5=Gabr |first5=Mohamed |last6=Matsushita |first6=Hitokazu |last7=Kim |first7=Young Jin |last8=Afify |first8=Mohamed |last9=Awadalla |first9=Hany |title=How Good Are GPT Models at Machine Translation? A Comprehensive Evaluation |date=2023-02-18 |eprint=2302.09210 |class=cs.CL}}</ref><ref>{{cite news |last1=Fadelli |first1=Ingrid |title=Study assesses the quality of AI literary translations by comparing them with human translations |url=https://techxplore.com/news/2022-11-quality-ai-literary-human.html |access-date=18 December 2022 |work=techxplore.com |language=en}}</ref><ref name="arxiv221014250">{{Cite arXiv|last1=Thai |first1=Katherine |last2=Karpinska |first2=Marzena |last3=Krishna |first3=Kalpesh |last4=Ray |first4=Bill |last5=Inghilleri |first5=Moira |last6=Wieting |first6=John |last7=Iyyer |first7=Mohit |title=Exploring Document-Level Literary Machine Translation with Parallel Paragraphs from World Literature |date=25 October 2022|class=cs.CL |eprint=2210.14250 }}</ref> This approach is considered promising,<ref name="WMT2023">{{cite conference |last1=Kocmi |first1=Tom |last2=Avramidis |first2=Eleftherios |last3=Bawden |first3=Rachel |last4=Bojar |first4=Ondřej |last5=Dvorkovich |first5=Anton |last6=Federmann |first6=Christian |last7=Fishel |first7=Mark |last8=Freitag |first8=Markus |last9=Gowda |first9=Thamme |last10=Grundkiewicz |first10=Roman |last11=Haddow |first11=Barry |last12=Koehn |first12=Philipp |last13=Marie |first13=Benjamin |last14=Monz |first14=Christof |last15=Morishita |first15=Makoto |date=2023 |editor-last=Koehn |editor-first=Philipp |editor2-last=Haddow |editor2-first=Barry |editor3-last=Kocmi |editor3-first=Tom |editor4-last=Monz |editor4-first=Christof |title=Findings of the 2023 Conference on Machine Translation (WMT23): LLMs Are Here but Not Quite There Yet |url=https://aclanthology.org/2023.wmt-1.1 |journal=Proceedings of the Eighth Conference on Machine Translation |location=Singapore |publisher=Association for Computational Linguistics |pages=1–42 |doi=10.18653/v1/2023.wmt-1.1|doi-access=free }}</ref> but is still more resource-intensive than specialized translation models.

==Issues==
[[File:Stir Fried Wikipedia.jpg|thumb|right|250px|A bilingual menu in [[Beijing]] offers "stir-fried wikipedia with [[Pimiento|pimientos]]" and other nonsensical dishes that may have resulted from a translation error. ]]
[[File:Machine translation in Bali.jpg|thumb|right|250px|Broken Chinese "{{lang|zh|沒有進入}}" from machine translation in [[Bali, Indonesia]]. The broken Chinese sentence sounds like "there does not exist an entry" or "have not entered yet".]]
Studies using human evaluation (e.g. by professional literary translators or human readers) have [[problem-solving|systematically identified various issues]] with the latest advanced MT outputs.<ref name="arxiv221014250"/> Some quality evaluation studies have found that, in several languages, human translations outperform ChatGPT-produced translations in terminological accuracy and clarity of expression.<ref>{{Cite journal |last1=Al Rousan |first1=Rafat |last2=Jaradat |first2=Raghad |last3=Malkawi |first3=Mona |date=2025-12-31 |title=ChatGPT translation vs. human translation: an examination of a literary text |journal=Cogent Social Sciences |volume=11 |issue=1 |article-number=2472916 |doi=10.1080/23311886.2025.2472916|doi-access=free }}</ref><ref>{{Cite journal |last1=Altakhaineh |first1=Abdel Rahman Mitib |last2=Alghathian |first2=Ghazi Ayed |last3=Jarrah |first3=Mashal Mufleh |date=2025-04-05 |title=A Comparative Study of Accuracy in Human vs. AI Translation of Legal Documents into Arabic |url=https://www.languageandlaw.eu/jll/article/view/190 |journal=International Journal of Language & Law (JLL) |language=en |volume=14 |pages=63–80 |issn=2194-7414}}</ref> Common issues include the translation of ambiguous parts whose correct translation requires common sense-like semantic language processing or context.<ref name="arxiv221014250"/> There can also be errors in the source texts, missing high-quality training data and the severity of frequency of several types of problems may not get reduced with techniques used to date, requiring some level of human active participation.

===Disambiguation===
{{Main|Word-sense disambiguation|Syntactic disambiguation}}
Word-sense disambiguation concerns finding a suitable translation when a word can have more than one meaning. The problem was first raised in the 1950s by [[Yehoshua Bar-Hillel]].<ref>[http://ourworld.compuserve.com/homepages/WJHutchins/Miles-6.htm Milestones in machine translation – No.6: Bar-Hillel and the nonfeasibility of FAHQT] {{webarchive|url=https://web.archive.org/web/20070312062051/http://ourworld.compuserve.com/homepages/WJHutchins/Miles-6.htm |date=12 March 2007 }} by John Hutchins</ref> He pointed out that without a "universal encyclopedia", a machine would never be able to distinguish between the two meanings of a word.<ref>Bar-Hillel (1960), "Automatic Translation of Languages". Available online at http://www.mt-archive.info/Bar-Hillel-1960.pdf {{Webarchive|url=https://web.archive.org/web/20110928112348/http://www.mt-archive.info/Bar-Hillel-1960.pdf |date=28 September 2011 }}</ref> Today there are numerous approaches designed to overcome this problem. They can be approximately divided into "shallow" approaches and "deep" approaches.

Shallow approaches assume no knowledge of the text. They simply apply statistical methods to the words surrounding the ambiguous word. Deep approaches presume a comprehensive knowledge of the word. So far, shallow approaches have been more successful.<ref>{{Cite book|title=Hybrid approaches to machine translation|author1=Costa-jussà, Marta R.|author2=Rapp, Reinhard|author3=Lambert, Patrik|author4=Eberle, Kurt|author5=Banchs, Rafael E.|author6=Babych, Bogdan|date=21 July 2016|isbn=978-3-319-21310-1|publisher=Springer International Publishing|location=Switzerland|oclc=953581497}}</ref>

[[Claude Piron]], a long-time translator for the United Nations and the [[World Health Organization]], wrote that machine translation, at its best, automates the easier part of a translator's job; the harder and more time-consuming part usually involves doing extensive research to resolve [[ambiguity|ambiguities]] in the [[source text]], which the [[grammatical]] and [[Lexical (semiotics)|lexical]] exigencies of the [[Translation|target language]] require to be resolved:

{{Blockquote|Why does a translator need a whole workday to translate five pages, and not an hour or two? ..... About 90% of an average text corresponds to these simple conditions.  But unfortunately, there's the other 10%.  It's that part that requires six [more] hours of work.  There are ambiguities one has to resolve.  For instance, the author of the source text, an Australian physician, cited the example of an epidemic which was declared during World War II in a "Japanese prisoners of war camp".  Was he talking about an American camp with Japanese prisoners or a Japanese camp with American prisoners?  The English has two senses.  It's necessary therefore to do research, maybe to the extent of a phone call to Australia.<ref name="piron">[[Claude Piron]], ''Le défi des langues'' (The Language Challenge), Paris, L'Harmattan, 1994. <!-- GFDL translation by Jim Henry --></ref>
}}

The ideal deep approach would require the translation software to do all the research necessary for this kind of disambiguation on its own; but this would require a higher degree of [[AI]] than has yet been attained.  A shallow approach which simply guessed at the sense of the ambiguous English phrase that Piron mentions (based, perhaps, on which kind of prisoner-of-war camp is more often mentioned in a given corpus) would have a reasonable chance of guessing wrong fairly often.  A shallow approach that involves "ask the user about each ambiguity" would, by Piron's estimate, only automate about 25% of a professional translator's job, leaving the harder 75% still to be done by a human.

===Non-standard speech===
One of the major pitfalls of MT is its inability to translate non-standard language with the same accuracy as standard language. Heuristic or statistical based MT takes input from various sources in standard form of a language. Rule-based translation, by nature, does not include common non-standard usages. This causes errors in translation from a vernacular source or into colloquial language. Limitations on translation from casual speech present issues in the use of machine translation in mobile devices.

===Named entities===
{{main|Named entity}}
In [[information extraction]], named entities, in a narrow sense, refer to concrete or abstract entities in the real world such as people, organizations, companies, and places that have a proper name: George Washington, Chicago, Microsoft.  It also refers to expressions of time, space and quantity such as 1 July 2011, $500.

In the sentence "Smith is the president of Fabrionix" both ''Smith'' and ''Fabrionix'' are named entities, and can be further qualified via first name or other information; "president" is not, since Smith could have earlier held another position at Fabrionix, e.g. Vice President.
The term [[rigid designator]] is what defines these usages for analysis in statistical machine translation.

Named entities must first be identified in the text; if not, they may be erroneously translated as common nouns, which would most likely not affect the [[Bilingual evaluation understudy|BLEU]] rating of the translation but would change the text's human readability.<ref>{{Cite conference |last1=Babych |first1=Bogdan |last2=Hartley |first2=Anthony |date=2003 |title=Improving Machine Translation Quality with Automatic Named Entity Recognition |url=http://www.cl.cam.ac.uk/~ar283/eacl03/workshops03/W03-w1_eacl03babych.local.pdf |conference=Paper presented at the 7th International EAMT Workshop on MT and Other Language Technology Tools... |archive-url=https://web.archive.org/web/20060514031411/http://www.cl.cam.ac.uk/~ar283/eacl03/workshops03/W03-w1_eacl03babych.local.pdf |archive-date=14 May 2006 |access-date=4 November 2013 }}</ref> They may be omitted from the output translation, which would also have implications for the text's readability and message.

[[Transliteration]] includes finding the letters in the target language that most closely correspond to the name in the source language.  This, however, has been cited as sometimes worsening the quality of translation.<ref>Hermajakob, U., Knight, K., & Hal, D. (2008).  [http://www.aclweb.org/old_anthology/P/P08/P08-1.pdf#page=433 Name Translation in Statistical Machine Translation Learning When to Transliterate] {{Webarchive|url=https://web.archive.org/web/20180104073326/http://www.aclweb.org/old_anthology/P/P08/P08-1.pdf#page=433 |date=4 January 2018 }}.  Association for Computational Linguistics.  389–397.</ref> For "Southern California" the first word should be translated directly, while the second word should be transliterated.  Machines often transliterate both because they treated them as one entity.  Words like these are hard for machine translators, even those with a transliteration component, to process.

Use of a "do-not-translate" list, which has the same end goal – transliteration as opposed to translation.<ref name="singla">{{Citation |last1=Neeraj Agrawal |title=Using Named Entity Recognition to improve Machine Translation |url=http://nlp.stanford.edu/courses/cs224n/2010/reports/singla-nirajuec.pdf |archive-url=https://web.archive.org/web/20130521075940/http://nlp.stanford.edu/courses/cs224n/2010/reports/singla-nirajuec.pdf |access-date=4 November 2013 |archive-date=21 May 2013 |last2=Ankush Singla |mode=cs1 |url-status=live}}</ref>  still relies on correct identification of named entities.

A third approach is a class-based model. Named entities are replaced with a token to represent their "class"; "Ted"  and "Erica" would both be replaced with "person" class token. Then the statistical distribution and use of person names, in general, can be analyzed instead of looking at the distributions of "Ted" and "Erica" individually, so that the probability of a given name in a specific language will not affect the assigned probability of a translation. A study by Stanford on improving this area of translation gives the examples that different probabilities will be assigned to "David is going for a walk" and "Ankit is going for a walk" for English as a target language due to the different number of occurrences for each name in the training data. A frustrating outcome of the same study by Stanford (and other attempts to improve named recognition translation) is that many times, a decrease in the [[Bilingual evaluation understudy|BLEU]] scores for translation will result from the inclusion of methods for named entity translation.<ref name="singla" />

==Applications==
While no system provides the ideal of fully automatic high-quality machine translation of unrestricted text, many fully automated systems produce reasonable output.<ref>{{Cite book |url=http://www.benjamins.com/cgi-bin/t_bookview.cgi?bookid=BTL%2014 |title=Melby, Alan. The Possibility of Language (Amsterdam:Benjamins, 1995, 27–41) |publisher=Benjamins.com |year=1995 |isbn=978-90-272-1614-4 |access-date=2012-06-12 |archive-url=https://web.archive.org/web/20110525234319/http://www.benjamins.com/cgi-bin/t_bookview.cgi?bookid=BTL%2014 |archive-date=25 May 2011 |url-status=live}}</ref><ref>{{Cite web |last=Wooten |first=Adam |date=14 February 2006 |title=A Simple Model Outlining Translation Technology |url=https://tandibusiness.blogspot.com/2006/02/simple-model-outlining-translation.html |archive-url=https://archive.today/20120716095630/http://tandibusiness.blogspot.de/2006/02/simple-model-outlining-translation.html |archive-date=16 July 2012 |access-date=2012-06-12 |website=T&I Business}}</ref><ref>{{cite web |url=http://www.mt-archive.info/Bar-Hillel-1960-App3.pdf |title=Appendix III of 'The present status of automatic translation of languages', Advances in Computers, vol.1 (1960), p.158-163. Reprinted in Y.Bar-Hillel: Language and information (Reading, Mass.: Addison-Wesley, 1964), p.174-179. |access-date=2012-06-12 |archive-date=28 September 2018 |archive-url=https://web.archive.org/web/20180928203641/http://www.mt-archive.info/Bar-Hillel-1960-App3.pdf }}</ref> The quality of machine translation is substantially improved if the domain is restricted and controlled.<ref>{{cite web |url=http://tauyou.com/blog/?p=47 |title=Human quality machine translation solution by Ta with you |language=es |publisher=Tauyou.com |date=15 April 2009 |access-date=2012-06-12 |archive-date=22 September 2009 |archive-url=https://web.archive.org/web/20090922094140/http://tauyou.com/blog/?p=47 |url-status=live }}</ref> This enables using machine translation as a tool to speed up and simplify translations, as well as producing flawed but useful low-cost or ad-hoc translations.

===Travel===
Machine translation applications have also been released for most mobile devices, including mobile telephones, pocket PCs, PDAs, etc. Due to their portability, such instruments have come to be designated as [[mobile translation]] tools enabling mobile business networking between partners speaking different languages, or facilitating both foreign language learning and unaccompanied traveling to foreign countries without the need of the intermediation of a human translator.

For example, the Google Translate app allows foreigners to quickly translate text in their surrounding via [[augmented reality]] using the smartphone camera that overlays the translated text onto the text.<ref>{{cite news |title=Google Translate Adds 20 Languages To Augmented Reality App |url=https://www.popsci.com/google-translate-adds-augmented-reality-translation-app/ |access-date=9 January 2023 |work=Popular Science |date=30 July 2015}}</ref> It can also [[speech recognition|recognize speech]] and then translate it.<ref>{{cite news |last1=Whitney |first1=Lance |title=Google Translate app update said to make speech-to-text even easier |url=https://www.cnet.com/tech/services-and-software/google-translation-app-may-better-recognize-certain-languages/ |access-date=9 January 2023 |work=CNET |language=en}}</ref>

===Public administration===
Despite their inherent limitations, MT programs are used around the world. Probably the largest institutional user is the [[European Commission]]. In 2012, with an aim to replace a rule-based MT by newer, statistical-based MT@EC, The European Commission contributed 3.072 million euros (via its ISA programme).<ref>{{cite web|url=http://ec.europa.eu/isa/actions/02-interoperability-architecture/2-8action_en.htm|title=Machine Translation Service|date=5 August 2011|access-date=13 September 2013|archive-date=8 September 2013|archive-url=https://web.archive.org/web/20130908232212/http://ec.europa.eu/isa/actions/02-interoperability-architecture/2-8action_en.htm|url-status=live}}</ref>

===Wikipedia===
Machine translation has also been used for translating [[Wikipedia]] articles and could play a larger role in creating, updating, expanding, and generally improving articles in the future, especially as the MT capabilities may improve. There is a "content translation tool" which allows editors to more easily translate articles across several select languages.<ref>{{cite news |last1=Wilson |first1=Kyle |title=Wikipedia has a Google Translate problem |url=https://www.theverge.com/2019/5/8/18526739/wikipedia-translation-tool-machine-learning-ai-english |access-date=9 January 2023 |work=The Verge |date=8 May 2019}}</ref><ref>{{cite news |title=Wikipedia taps Google to help editors translate articles |url=https://venturebeat.com/ai/wikipedia-taps-google-to-help-editors-translate-articles/ |access-date=9 January 2023 |work=VentureBeat |date=9 January 2019}}</ref><ref>{{cite web |title=Content translation tool helps create over half a million Wikipedia articles |url=https://wikimediafoundation.org/news/2019/09/23/content-translation-tool-helps-create-over-half-a-million-wikipedia-articles/ |website=Wikimedia Foundation |access-date=10 January 2023 |date=23 September 2019}}</ref> English-language articles are thought to usually be more comprehensive and less biased than their non-translated equivalents in other languages.<ref>{{cite web |last1=Magazine |first1=Undark |title=Wikipedia Has a Language Problem. Here's How To Fix It. |url=https://undark.org/2021/08/12/wikipedia-has-a-language-problem-heres-how-to-fix-it/ |website=Undark Magazine |access-date=9 January 2023 |date=12 August 2021}}</ref> As of 2022, [[English Wikipedia]] has over 6.5 million articles while, for example, the [[German Wikipedia|German]] and [[Swedish Wikipedia]]s each only have over 2.5 million articles,<ref>{{cite web |title=List of Wikipedias - Meta |url=https://meta.wikimedia.org/wiki/List_of_Wikipedias |website=meta.wikimedia.org |access-date=9 January 2023 |language=en}}</ref> each often far less comprehensive.

===Surveillance and military===
Following terrorist attacks in Western countries, including [[9-11]], the U.S. and its allies have been most interested in developing  [[Arabic machine translation]] programs, but also in translating [[Pashto language|Pashto]] and [[Dari (Eastern Persian)|Dari]] languages.{{Citation needed|date=February 2007}} Within these languages, the focus is on key phrases and quick communication between military members and civilians through the use of mobile phone apps.<ref>{{cite journal |last=Gallafent |first=Alex |title=Machine Translation for the Military |journal=PRI's the World |date=26 Apr 2011 |access-date=17 Sep 2013 |url=http://www.theworld.org/2011/04/machine-translation-military/ |archive-date=9 May 2013 |archive-url=https://web.archive.org/web/20130509171415/http://www.theworld.org/2011/04/machine-translation-military/ |url-status=live }}</ref> The Information Processing Technology Office in [[DARPA]] hosted programs like [[DARPA TIDES program|TIDES]] and [[Babylon translator]]. US Air Force has awarded a $1 million contract to develop a language translation technology.<ref>{{cite web |last=Jackson |first=William |url=http://gcn.com/articles/2003/09/09/air-force-wants-to-build-a-universal-translator.aspx |title=GCN – Air force wants to build a universal translator |publisher=Gcn.com |date=9 September 2003 |access-date=2012-06-12 |archive-date=16 June 2011 |archive-url=https://web.archive.org/web/20110616052943/http://gcn.com/articles/2003/09/09/air-force-wants-to-build-a-universal-translator.aspx |url-status=live }}</ref>

===Social media===
The notable rise of [[social networking]] on the web in recent years has created yet another niche for the application of machine translation software – in utilities such as [[Facebook]], or [[instant messaging]] clients such as [[Skype]], [[Google Talk]], [[MSN Messenger]], etc. – allowing users speaking different languages to communicate with each other.

==== Online games ====
[[Lineage W]] gained popularity in Japan because of its machine translation features allowing players from different countries to communicate.<ref>{{Cite web |last=Young-sil |first=Yoon |date=2023-06-26 |title=Korean Games Growing in Popularity in Tough Japanese Game Market |url=http://www.businesskorea.co.kr/news/articleView.html?idxno=117105 |access-date=2023-08-08 |website=BusinessKorea |language=}}</ref>

===Medicine===
Despite being labelled as an unworthy competitor to human translation in 1966 by the Automated Language Processing Advisory Committee put together by the United States government,<ref>{{Cite report |url=http://www.nap.edu/html/alpac_lm/ARC000005.pdf |title=Language and Machines: Computers in Translation and Linguistics |author=((Automatic Language Processing Advisory Committee, Division of Behavioral Sciences, National Academy of Sciences, National Research Council)) |date=1966 |publisher=National Research Council, National Academy of Sciences |location=Washington, D. C. |access-date=21 October 2013 |archive-url=https://web.archive.org/web/20131021044934/http://www.nap.edu/html/alpac_lm/ARC000005.pdf |archive-date=21 October 2013 |url-status=live}}</ref> the quality of machine translation has now been improved to such levels that its application in online collaboration and in the medical field are being investigated. The application of this technology in medical settings where human translators are absent is another topic of research, but difficulties arise due to the importance of accurate translations in medical diagnoses.<ref>{{cite journal|url=http://www.cfp.ca/content/59/4/382.full|title=Using machine translation in clinical practice|journal=Canadian Family Physician|date=April 2013|volume=59|issue=4|pages=382–383|access-date=21 October 2013|archive-date=4 May 2013|archive-url=https://web.archive.org/web/20130504040732/http://www.cfp.ca/content/59/4/382.full|url-status=live|last1=Randhawa|first1=Gurdeeshpal|last2=Ferreyra|first2=Mariella|last3=Ahmed|first3=Rukhsana|last4=Ezzat|first4=Omar|last5=Pottie|first5=Kevin|pmid=23585608|pmc=3625087}}</ref>

Researchers caution that the use of machine translation in medicine could risk mistranslations that can be dangerous in critical situations.<ref name=":02">{{Cite journal |last1=Vieira |first1=Lucas Nunes |last2=O'Hagan |first2=Minako |last3=O'Sullivan |first3=Carol |date=2021-08-18 |title=Understanding the societal impacts of machine translation: a critical review of the literature on medical and legal use cases |journal=Information, Communication & Society |language=en |volume=24 |issue=11 |pages=1515–1532 |doi=10.1080/1369118X.2020.1776370 |s2cid=225694304 |issn=1369-118X|doi-access=free |hdl=1983/29727bd1-a1ae-4600-9e8e-018f11ec75fb |hdl-access=free }}</ref><ref>{{Cite journal |last1=Khoong |first1=Elaine C. |last2=Steinbrook |first2=Eric |last3=Brown |first3=Cortlyn |last4=Fernandez |first4=Alicia |date=2019-04-01 |title=Assessing the Use of Google Translate for Spanish and Chinese Translations of Emergency Department Discharge Instructions |journal=JAMA Internal Medicine |language=en |volume=179 |issue=4 |pages=580–582 |doi=10.1001/jamainternmed.2018.7653 |issn=2168-6106 |pmc=6450297 |pmid=30801626}}</ref> Machine translation can make it easier for doctors to communicate with their patients in day to day activities, but it is recommended to only use machine translation when there is no other alternative, and that translated medical texts should be reviewed by human translators for accuracy.<ref>{{Cite journal |last=Piccoli |first=Vanessa |date=2022-07-05 |title=Plurilingualism, multimodality and machine translation in medical consultations: A case study |url=http://www.jbe-platform.com/content/journals/10.1075/tis.21012.pic |journal=Translation and Interpreting Studies |language=en |volume=17 |issue=1 |pages=42–65 |doi=10.1075/tis.21012.pic |s2cid=246780731 |issn=1932-2798|url-access=subscription }}</ref><ref>{{Cite journal |last1=Herrera-Espejel |first1=Paula Sofia |last2=Rach |first2=Stefan |date=2023-11-20 |title=The Use of Machine Translation for Outreach and Health Communication in Epidemiology and Public Health: Scoping Review |journal=JMIR Public Health and Surveillance |language=en |volume=9 |article-number=e50814 |doi=10.2196/50814 |pmid=37983078 |issn=2369-2960 |doi-access=free |pmc=10696499 }}</ref>

=== Law ===
[[Legal English|Legal language]] poses a significant challenge to machine translation tools due to its precise nature and atypical use of normal words. For this reason, specialized algorithms have been developed for use in legal contexts.<ref name=":1">{{Cite web |last=legalj |date=2023-01-02 |title=Man v. Machine: Social and Legal Implications of Machine Translation |url=https://legaljournal.princeton.edu/man-v-machine-social-and-legal-implications-of-machine-translation/ |access-date=2023-12-04 |website=Princeton Legal Journal |language=en-US}}</ref> Due to the risk of mistranslations arising from machine translators, researchers recommend that machine translations should be reviewed by human translators for accuracy, and some courts prohibit its use in [[Legal proceeding|formal proceedings]].<ref>{{Cite journal |last=Chavez |first=Edward L. |date=2008 |title=New Mexico's Success with Non-English Speaking Jurors |url=https://heinonline.org/HOL/Page?handle=hein.journals/jrlci1&id=307&div=&collection= |journal=Journal of Court Innovation |volume=1 |page=303}}</ref>

The use of machine translation in law has raised concerns about translation errors and [[client confidentiality]]. Lawyers who use free translation tools such as Google Translate may accidentally violate client confidentiality by exposing private information to the providers of the translation tools.<ref name=":1" /> In addition, there have been arguments that consent for a police search that is obtained with machine translation is invalid, with different courts issuing different verdicts over whether or not these arguments are valid.<ref name=":02"/>

=== Ancient languages ===
The advancements in [[convolutional neural network]]s in recent years and in low resource machine translation (when only a very limited amount of data and examples are available for training) enabled machine translation for ancient languages, such as [[Akkadian language|Akkadian]] and its dialects Babylonian and Assyrian.<ref>{{Cite journal |last1=Gutherz |first1=Gai |last2=Gordin |first2=Shai |last3=Sáenz |first3=Luis |last4=Levy |first4=Omer |last5=Berant |first5=Jonathan |date=2023-05-02 |editor-last=Kearns |editor-first=Michael |title=Translating Akkadian to English with neural machine translation |url=https://academic.oup.com/pnasnexus/article/doi/10.1093/pnasnexus/pgad096/7147349 |journal=PNAS Nexus |language=en |volume=2 |issue=5 |article-number=pgad096 |doi=10.1093/pnasnexus/pgad096  |doi-access=free|issn=2752-6542 |pmc=10153418 |pmid=37143863}}</ref>

==Evaluation==
{{Main|Evaluation of machine translation}}
There are many factors that affect how machine translation systems are evaluated. These factors include the intended use of the translation, the nature of the machine translation software, and the nature of the translation process.

Different programs may work well for different purposes. For example, [[statistical machine translation]] (SMT) typically outperforms [[example-based machine translation]] (EBMT), but researchers found that when evaluating English to French translation, EBMT performs better.<ref name="Way 295–309">{{cite journal|last=Way|first=Andy|author2=Nano Gough|title=Comparing Example-Based and Statistical Machine Translation|journal=Natural Language Engineering|date=20 September 2005|volume=11|issue=3|pages=295–309|doi=10.1017/S1351324905003888|s2cid=3242163}}</ref> The same concept applies for technical documents, which can be more easily translated by SMT because of their formal language.

In certain applications, however, e.g., product descriptions written in a [[controlled language]], a [[dictionary-based machine translation|dictionary-based machine-translation]] system has produced satisfactory translations that require no human intervention save for quality inspection.<ref>Muegge (2006), "[http://www.mt-archive.info/Aslib-2006-Muegge.pdf Fully Automatic High Quality Machine Translation of Restricted Text: A Case Study] {{Webarchive|url=https://web.archive.org/web/20111017043848/http://www.mt-archive.info/Aslib-2006-Muegge.pdf |date=17 October 2011 }}," in ''Translating and the computer 28. Proceedings of the twenty-eighth international conference on translating and the computer, 16–17 November 2006, London'', London: Aslib. {{ISBN|978-0-85142-483-5}}.</ref>

There are various means for evaluating the output quality of machine translation systems. The oldest is the use of human judges<ref>{{cite web |url=http://www.morphologic.hu/public/mt/2008/compare12.htm |title=Comparison of MT systems by human evaluation, May 2008 |publisher=Morphologic.hu |access-date=2012-06-12 |archive-url=https://web.archive.org/web/20120419072313/http://www.morphologic.hu/public/mt/2008/compare12.htm |archive-date=19 April 2012 }}</ref> to assess a translation's quality. Even though human evaluation is time-consuming, it is still the most reliable method to compare different systems such as rule-based and statistical systems.<ref>Anderson, D.D. (1995). [http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.961.5377&rep=rep1&type=pdf Machine translation as a tool in second language learning] {{Webarchive|url=https://web.archive.org/web/20180104073518/http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.961.5377&rep=rep1&type=pdf |date=4 January 2018 }}. CALICO Journal. 13(1). 68–96.</ref> [[Automate]]d means of evaluation include [[Bilingual evaluation understudy|BLEU]], [[NIST (metric)|NIST]], [[METEOR]], and [[LEPOR]].<ref>Han et al. (2012), "[http://repository.umac.mo/jspui/bitstream/10692/1747/1/10205_0_%5B2012-12-08~15%5D%20C.%20%28COLING2012%29%20LEPOR.pdf LEPOR: A Robust Evaluation Metric for Machine Translation with Augmented Factors] {{Webarchive|url=https://web.archive.org/web/20180104073506/http://repository.umac.mo/jspui/bitstream/10692/1747/1/10205_0_%5B2012-12-08~15%5D%20C.%20%28COLING2012%29%20LEPOR.pdf |date=4 January 2018 }}," in ''Proceedings of the 24th International Conference on Computational Linguistics (COLING 2012): Posters, pages 441–450'', Mumbai, India.</ref>

Relying exclusively on unedited machine translation ignores the fact that communication in [[natural language|human language]] is context-embedded and that it takes a person to comprehend the [[context (language use)|context]] of the original text with a reasonable degree of probability. It is certainly true that even purely human-generated translations are prone to error. Therefore, to ensure that a machine-generated translation will be useful to a human being and that publishable-quality translation is achieved, such translations must be reviewed and edited by a human.<ref>J.M. Cohen observes (p.14): "Scientific translation is the aim of an age that would reduce all activities to [[Technology|techniques]]. It is impossible however to imagine a literary-translation machine less complex than the human brain itself, with all its knowledge, reading, and discrimination."</ref> The late [[Claude Piron]] wrote that machine translation, at its best, automates the easier part of a translator's job; the harder and more time-consuming part usually involves doing extensive research to resolve [[ambiguity|ambiguities]] in the [[source text]], which the [[grammatical]] and [[Lexical (semiotics)|lexical]] exigencies of the target language require to be resolved. Such research is a necessary prelude to the pre-editing necessary in order to provide input for machine-translation software such that the output will not be [[garbage in garbage out|meaningless]].<ref name="NIST">See the [https://www.nist.gov/speech/tests/mt/ annually performed NIST tests since 2001] {{Webarchive|url=https://web.archive.org/web/20090322202656/http://nist.gov/speech/tests/mt/ |date=22 March 2009 }} and [[Bilingual Evaluation Understudy]]</ref>

In addition to disambiguation problems, decreased accuracy can occur due to varying levels of training data for machine translating programs. Both example-based and statistical machine translation rely on a vast array of real example sentences as a base for translation, and when too many or too few sentences are analyzed accuracy is jeopardized. Researchers found that when a program is trained on 203,529 sentence pairings, accuracy actually decreases.<ref name="Way 295–309"/> The optimal level of training data seems to be just over 100,000 sentences, possibly because as training data increases, the number of possible sentences increases, making it harder to find an exact translation match.

Flaws in machine translation have been noted for [[Humour in translation|their entertainment value]]. Two videos uploaded to [[YouTube]] in April 2017 involve two Japanese [[hiragana]] characters えぐ (''[[E (kana)|e]]'' and ''[[Ku (kana)|gu]]'') being repeatedly pasted into Google Translate, with the resulting translations quickly degrading into nonsensical phrases such as "DECEARING EGG" and "Deep-sea squeeze trees", which are then read in increasingly absurd voices;<ref>{{Cite web|url=https://www.businessinsider.com/google-translate-fails-2017-11|title=4 times Google Translate totally dropped the ball|first=Mark|last=Abadi|website=Business Insider}}</ref><ref>{{Cite web|url=https://nlab.itmedia.co.jp/nl/articles/1704/16/news013.html|title=回数を重ねるほど狂っていく Google翻訳で「えぐ」を英訳すると奇妙な世界に迷い込むと話題に|website=ねとらぼ |date=16 April 2017 }}</ref> the full-length version of the video currently has 7.1 million views {{as of|lc=y|August 2025|post=.}}<ref>{{Cite web|url=https://www.youtube.com/watch?v=3-rfBsWmo0M|title=えぐ|date=12 April 2017 |via=www.youtube.com}}</ref>

==Machine translation and signed languages==
{{main|Machine translation of sign languages}}
In the early 2000s, options for machine translation between spoken and signed languages were severely limited. It was a common belief that deaf individuals could use traditional translators. However, stress, intonation, pitch, and timing are conveyed much differently in spoken languages compared to signed languages. Therefore, a deaf individual may misinterpret or become confused about the meaning of written text that is based on a spoken language.<ref name="Zhao, L. 2000">Zhao, L., Kipper, K., Schuler, W., Vogler, C., & Palmer, M. (2000). [http://repository.upenn.edu/cgi/viewcontent.cgi?article=1043&context=hms A Machine Translation System from English to American Sign Language] {{Webarchive|url=https://web.archive.org/web/20180720012839/https://repository.upenn.edu/cgi/viewcontent.cgi?article=1043&context=hms |date=20 July 2018 }}. Lecture Notes in Computer Science, 1934: 54–67.</ref>

Researchers Zhao, et al. (2000), developed a prototype called TEAM (translation from English to ASL by machine) that completed English to [[American Sign Language]] (ASL) translations. The program would first analyze the syntactic, grammatical, and morphological aspects of the English text. Following this step, the program accessed a sign synthesizer, which acted as a dictionary for ASL. This synthesizer housed the process one must follow to complete ASL signs, as well as the meanings of these signs. Once the entire text is analyzed and the signs necessary to complete the translation are located in the synthesizer, a computer generated human appeared and would use ASL to sign the English text to the user.<ref name="Zhao, L. 2000"/>

==Copyright==
Only [[creative work|work]]s that are [[originality|original]] are subject to [[copyright]] protection, so some scholars claim that machine translation results are not entitled to copyright protection because MT does not involve [[creativity]].<ref>{{cite web|url=http://www.seo-translator.com/machine-translation-no-copyright-on-the-result/|title=Machine Translation: No Copyright On The Result?|access-date=24 November 2012|publisher=SEO Translator, citing [[Zimbabwe Independent]]|archive-date=29 November 2012|archive-url=https://web.archive.org/web/20121129042959/http://www.seo-translator.com/machine-translation-no-copyright-on-the-result/|url-status=live}}</ref> The copyright at issue is for a [[derivative work]]; the author of the [[originality|original work]] in the original language does not lose his [[rights]] when a work is translated: a translator must have permission to [[publishing|publish]] a translation.<ref>{{Cite web |title=translation {{!}} UCL Copyright Queries |url=https://blogs.ucl.ac.uk/copyright/tag/translation/ |access-date=2025-08-18 |website=blogs.ucl.ac.uk}}</ref>

==See also==
{{div col|colwidth=18em}}
*[[AI-complete]]
*[[Cache language model]]
*[[Comparison of machine translation applications]]
*[[Comparison of different machine translation approaches]]
*[[Computational linguistics]]
*[[Computer-assisted translation]] and [[Translation memory]]
*[[Controlled language in machine translation]]
*[[Controlled natural language]]
*[[Foreign language writing aid]]
*[[Fuzzy matching (computer-assisted translation)]]
*[[History of machine translation]]
*[[Human language technology]]
*[[Humour in translation]] ("howlers")
*[[Language and Communication Technologies]]
*[[Language barrier]]
*[[List of emerging technologies]]
*[[List of research laboratories for machine translation]]
*[[Mobile translation]]
*[[Neural machine translation]]
*[[OpenLogos]]
*[[Phraselator]]
*[[Postediting]]
*[[Pseudo-translation]]
*[[Round-trip translation]]
*[[Statistical machine translation]]
*{{section link|Translation#Machine translation}}
*[[Translation memory]]
*[[ULTRA (machine translation system)]]
*[[Universal Networking Language]]
*[[Universal translator]]
{{div col end}}

==Notes==
{{reflist}}

==Further reading==
* {{cite book |editor1-last=Nirenburg |editor1-first=Sergei |editor2-last=Somers |editor2-first=Harold L. |editor3-last=Wilks |editor3-first=Yorick A. |title=Readings in Machine Translation |date=2003 |publisher=The MIT Press |location=Cambridge, MA |isbn=9780262140744}}
* {{Citation |last=Cohen |first=J. M. |contribution=Translation |title=Encyclopedia Americana |year=1986 |volume=27 |pages=12–15 |ref=none|title-link=Encyclopedia Americana }}
* {{Cite book |last1=Hutchins |first1=W. John |url=https://archive.org/details/introductiontoma0000hutc |title=An Introduction to Machine Translation |last2=Somers |first2=Harold L. |publisher=Academic Press |year=1992 |isbn=0-12-362830-X |location=London |author-link=W. John Hutchins |url-access=registration}}
* {{Cite magazine |last=Lewis-Kraus |first=Gideon |date=7 June 2015 |title=Tower of Babble |magazine=New York Times Magazine |pages=48–52}}
* {{Cite journal |last1=Weber |first1=Steven |last2=Mehandru |first2=Nikita |date=2022 |title=The 2020s Political Economy of Machine Translation |journal=Business and Politics |language=en |volume=24 |issue=1 |pages=96–112 |doi=10.1017/bap.2021.17|arxiv=2011.01007 |s2cid=226236853 }}
* {{cite book |last=Eisenstein  |first=Jacob |date=October 1, 2019 |title=Introduction to Natural Language Processing |url=https://mitpress.mit.edu/9780262042840/introduction-to-natural-language-processing/ |location= |publisher=The MIT Press |page=431-456 |isbn=978-0-262-04284-0 |chapter=Machine translation}}

==External links==
{{Wikiversity|Topic:Computational linguistics}}
* [http://www.omniglot.com/language/articles/machinetranslation.htm The Advantages and Disadvantages of Machine Translation]
* [http://www.eamt.org/iamt.php International Association for Machine Translation (IAMT)] {{Webarchive|url=https://web.archive.org/web/20100624162302/http://www.eamt.org/iamt.php |date=24 June 2010 }}
*[http://www.mt-archive.info Machine Translation Archive] {{Webarchive|url=https://web.archive.org/web/20190401232615/http://www.mt-archive.info/ |date=1 April 2019 }} by [[W. John Hutchins|John Hutchins]]. An electronic repository (and bibliography) of articles, books and papers in the field of machine translation and computer-based translation technology
*[http://www.hutchinsweb.me.uk/ Machine translation (computer-based translation)] – Publications by John Hutchins (includes [[PDF format|PDF]]s of several books on machine translation)
*[https://web.archive.org/web/20080120192259/http://bowland-files.lancs.ac.uk/monkey/ihe/mille/paper2.htm Machine Translation and Minority Languages]
*[http://www.foreignword.com/Technology/art/Hutchins/hutchins99.htm John Hutchins 1999] {{Webarchive|url=https://web.archive.org/web/20070907182416/http://www.foreignword.com/Technology/art/Hutchins/hutchins99.htm |date=7 September 2007 }}
*[https://slator.com/machine-translation/ Slator News & analysis of the latest developments in machine translation]
*[https://www.machinetranslation.com/blog/how-machine-translation-is-changing-the-landscape-of-foreign-language-learning From Classroom to Real World: How Machine Translation is Changing the Landscape of Foreign Language Learning]
{{Natural Language Processing}}
{{Approaches to machine translation}}
{{emerging technologies|topics=yes|infocom=yes}}

{{Authority control}}

[[Category:Machine translation| ]]
[[Category:Applications of artificial intelligence]]
[[Category:Computational linguistics]]
[[Category:Computer-assisted translation]]
[[Category:Tasks of natural language processing]]
[[Category:Automation software]]