Comparison of machine translation applications
Machine translation is an algorithm which attempts to translate text or speech from one natural language to another.
01General information
Basic general information for popular machine translation applications.
| Name | Platform | License | Price | Latest stable release | Source availability | Translation via website | Number of supported languages | Notes |
|---|---|---|---|---|---|---|---|---|
| Anusaaraka | Unix compatible | GNU GPL | No fee required | 0.50 | Yes | Rule-based, deep parser based, paninian framework based; all programs and language data are free and open-source | ||
| Apertium | Cross-platform (web application), Unix compatible, precompiled packages available for Debian | GNU GPL | No fee required | 3.9.4 | Yes | Rule-based, shallow transfer; all programs and language data are free and open source | ||
| Babylon | Windows, Mac | Proprietary | Depends on license ($9.90-$89 for one license) | 11.0.1.6 | No | Prompts to install the Babylon Toolbar, a browser hijacker which is difficult to remove. | ||
| Baidu Fanyi | Cross-platform (web application) | SaaS | No fee required | No | 200+ | Neural machine translation | ||
| DeepL | Cross-platform (web application) | SaaS | No fee required, paid "Pro" version available | No | 120+ | Neural machine translation | ||
| Google Translate | Cross-platform (web application) | SaaS | No fee required | No | 240+ | Statistical and neural machine translation | ||
| GramTrans | Cross-platform (web application) | Free | No fee required | ? | No | Rule-based, using constraint grammar | ||
| IBM Watson | Cross-platform | SaaS | Free, commercial (varies by plan) | 3.0 | No | 55+ | Both rule-based and statistical models developed by IBM Research. Neural machine translation models available through the Watson Language Translator API for developers. | |
| Microsoft Translator | Cross-platform (web application) | SaaS | No fee required | No | 180+ | Statistical and neural machine translation | ||
| Moses | Cross-platform | GNU LGPL | No fee required | 4.0 | Yes | Drop-in replacement for Pharaoh, features factored translation models and decoding of confusion networks. | ||
| Moses for Mere Mortals | Ubuntu Linux | GNU GPL | No fee required | 2014-11-16 | Yes | Free open source; based on Moses; translation memories integration. | ||
| Naver Papago | Cross-platform (web application) | SaaS | No fee required | No | 15+ | |||
| NiuTrans | Cross-platform | GNU GPL | No fee required | 1.4.0 | Yes | 450+ | Competitive performance for Chinese translation tasks; statistical machine translation. Supports phrase-based, hierarchical phrase-based, and syntax-based (string-to-tree, tree-to-string, and tree-to-tree) models for research purposes. | |
| OpenLogos | Windows, Linux | GNU GPL or paid initiative taker | No fee required | 1.0.3 | Yes | Rule-based, deep transfer | ||
| PROMT | Cross-platform (web application) | Proprietary | Commercial | 12.5 | No | 25+ | Hybrid, rule-based, statistical and neural machine translation | |
| Reverso | Cross-platform (web application) | SaaS | No fee required, paid "Premium" version available | No | 30+ | Neural machine translation | ||
| SYSTRAN | Cross-platform (web application) | Proprietary | $200 (desktop), $15,000 and up (enterprise server) | Version 7 | No | 60+ | Hybrid, rule-based, statistical machine translation and neural machine translation | |
| Translate (Apple) | iOS, iPadOS and watchOS | Proprietary | ? | No | 30+ | Neural machine translation | ||
| Yandex.Translate | Cross-platform (web application) | SaaS | No fee required | No | 110+ | Statistical and neural machine translation |
02Languages features comparison
The following table compares the number of languages which the following machine translation programs can translate between. (Moses and Moses for Mere Mortals allow you to train translation models for any language pair, though collections of translated texts (parallel corpus) need to be provided by the user. The Moses site provides links to training corpora.)
This is not an all-encompassing list. Some applications have many more language pairs than those listed below. This is a general comparison of key languages only. A full and accurate list of language pairs supported by each product should be found on each of the product's websites.
Multi-pair translations
Paired translations
| Language | Language | Apertium | OpenLogos | GramTrans | SYSTRAN |
|---|---|---|---|---|---|
| Afrikaans | Dutch | ⇆ | No | No | No |
| Arabic | English | No | No | No | ⇆ |
| Arabic | Maltese | ← | No | No | No |
| Aragonese | Catalan | ⇆ | No | No | No |
| Aragonese | Spanish | ⇆ | No | No | No |
| Arpitan | French | ← | No | No | No |
| Asturian | Spanish | ← | No | No | No |
| Basque | English | → | No | No | No |
| Basque | Spanish | → | No | No | No |
| Belarusian | Russian | ⇆ | No | No | No |
| Breton | French | → | No | No | No |
| Bulgarian | Macedonian | ⇆ | No | No | No |
| Catalan | Danish | No | No | ← | No |
| Catalan | English | ⇆ | No | No | No |
| Catalan | Esperanto | → | No | No | No |
| Catalan | French | ⇆ | No | No | No |
| Catalan | Italian | ⇆ | No | No | No |
| Catalan | Occitan | ⇆ | No | No | No |
| Catalan | Portuguese | ⇆ | No | No | No |
| Catalan | Romanian | ⇆ | No | No | No |
| Catalan | Sardinian | ← | No | No | No |
| Catalan | Spanish | ⇆ | No | No | No |
| Chinese (Simplified) | English | No | No | No | ⇆ |
| Chinese (Traditional) | English | No | No | No | ⇆ |
| Crimean Tatar | Turkish | → | No | No | No |
| Czech | English | No | No | No | ⇆ |
| Danish | English | No | No | ⇆ | → |
| Danish | Esperanto | No | No | → | No |
| Danish | Galician | No | No | → | No |
| Danish | German | No | No | ⇆ | No |
| Danish | Norwegian (Bokmål) | ⇆ | No | ⇆ | No |
| Danish | Portuguese | No | No | ← | No |
| Danish | Spanish | No | No | ⇆ | No |
| Danish | Swedish | ⇆ | No | ⇆ | No |
| Dutch | English | No | No | No | ⇆ |
| Dutch | French | No | No | No | ⇆ |
| English | Esperanto | ⇆ | No | → | No |
| English | Finnish | No | No | No | ⇆ |
| English | French | No | ← | No | ⇆ |
| English | Galician | ⇆ | No | No | No |
| English | German | No | ⇆ | No | ⇆ |
| English | Greek | No | No | No | ⇆ |
| English | Hindi | No | No | No | ← |
| English | Hungarian | No | No | No | ⇆ |
| English | Icelandic | ← | No | No | No |
| English | Italian | No | ← | No | ⇆ |
| English | Japanese | No | No | No | ⇆ |
| English | Korean | No | No | No | ⇆ |
| English | Latvian | No | No | No | ← |
| English | Lithuanian | No | ← | No | No |
| English | Macedonian | ← | No | No | No |
| English | Norwegian | No | No | ⇆ | ← |
| English | Persian | No | No | No | ← |
| English | Polish | No | No | No | ⇆ |
| English | Portuguese | No | ← | ← | ⇆ |
| English | Romanian | No | No | No | ⇆ |
| English | Russian | No | ← | No | ⇆ |
| English | Serbo-Croatian | ← | No | No | No |
| English | Slovak | No | No | No | ← |
| English | Slovene | No | No | No | ← |
| English | Spanish | ⇆ | ← | No | ⇆ |
| English | Swedish | No | No | ⇆ | ⇆ |
| English | Turkish | No | No | No | ⇆ |
| English | Ukrainian | No | No | No | ⇆ |
| English | Welsh | ← | No | No | No |
| Esperanto | French | ← | No | No | No |
| Esperanto | Norwegian | No | No | ← | No |
| Esperanto | Portuguese | No | No | ← | No |
| Esperanto | Spanish | ← | No | No | No |
| Estonian | English | No | No | No | → |
| French | German | No | No | No | ⇆ |
| French | Italian | No | No | No | ⇆ |
| French | Occitan | ⇆ | No | No | No |
| French | Russian | No | ← | No | No |
| French | Spanish | ⇆ | No | No | ⇆ |
| Galician | Portuguese | ⇆ | No | No | No |
| Galician | Spanish | ⇆ | No | No | No |
| German | Italian | No | No | No | ⇆ |
| German | Spanish | No | No | No | ⇆ |
| Hindi | Urdu | ⇆ | No | No | No |
| Icelandic | Swedish | ⇆ | No | No | No |
| Indonesian | Malaysian | ⇆ | No | No | No |
| Italian | Sardinian | ⇆ | No | No | No |
| Italian | Spanish | No | No | No | ← |
| Kazakh | Tatar | ⇆ | No | No | No |
| Macedonian | Serbo-Croatian | ← | No | No | No |
| North Sámi | Norwegian | → | No | No | No |
| Norwegian | Swedish | ⇆ | No | ⇆ | No |
| Norwegian (Nynorsk) | Norwegian (Bokmål) | ⇆ | No | No | No |
| Occitan | Spanish | ⇆ | No | No | No |
| Polish | Silesian | → | No | No | No |
| Portuguese | Spanish | ⇆ | No | No | No |
| Romanian | Spanish | → | No | No | No |
| Russian | Ukrainian | ⇆ | No | No | No |
| Serbo-Croatian | Slovene | ⇆ | No | No | No |
| Total (confirmed) | 83 | 9 | 24 | 63 | |
Sources and credits
This article is adapted from the Wikipedia article “Comparison of machine translation applications”, written by its contributors and licensed under CC BY-SA 4.0. Fathomly has changed the layout, removed citation markers, navigation and maintenance notices, and adjusted punctuation. This adapted version is shared under the same license. For references, see the original article.
Fathomly is not affiliated with or endorsed by the Wikimedia Foundation. Spotted a problem? Tell us.