Microsoft Translator

from Wikipedia, the free encyclopedia
Globe icon of the infobox
Microsoft Translator
Machine translation
languages more than 60
On-line (currently online)
https://www.microsoft.com/translator/


Microsoft Translator is a multilingual machine translation cloud service from Microsoft. Microsoft Translator is integrated with several consumer, developer and business products including Bing , Microsoft Office , SharePoint , Microsoft Edge , Microsoft Lync , Yammer , Skype Translator, Visual Studio , Internet Explorer and Microsoft Translator applications for Windows, Windows Phone , iPhone and Apple Watch as well as Android Phone and Wear OS .

Microsoft Translator also offers text and language translations based on cloud services for companies. The service for text translation through the Translator Text API ranges from a free tier that supports two million characters per month to the paid tier that supports billions of characters per month.

The service supports more than 60 language systems (as of October 2019). It supports more than 10 language translation systems, which are the basis for Microsoft Translator Live Conversation, Skype and Skype for Windows Desktop and the Microsoft Translator Apps for iOS and Android.

history

The first version of Microsoft's machine translation system was developed between 1999 and 2000 as part of Microsoft Research . This system was based on semantic predicate argument structures, the so-called Logical Forms (LF), and was developed from the grammar correction developed for Microsoft Word . This system was eventually used to translate the entire Microsoft Knowledge Base into Spanish , French , German, and Japanese .

Microsoft's approach to machine translation, like most of today's modern machine translation systems, is "data-driven".

In 2011, a cloud- based API that supports products that are available to both home and business users, expanded the service with numerous Microsoft Translator products. In March 2016, an additional language translation function was introduced.

An update to the API was introduced in May 2018. This new version offers neural machine translation as the standard method of translation. In addition to translation, the new version offers transliteration and a bilingual dictionary to look up words, find alternative translations and show examples in sentences.

accuracy

The quality of Microsoft Translator's machine translation results is assessed using a method called the BLEU score.

BLEU (Bilingual Evaluation Understudy) is an algorithm for evaluating the quality of texts that have been translated from one natural language into another. Quality is understood as the consistency between the performance of a machine and that of a person. BLEU is one of the first metrics to achieve a high correlation with human quality judgments and is one of the most popular automated and inexpensive metrics.

Because machine translation is based on statistical algorithms rather than human translators, the automatic translations it produces are not always very accurate. Microsoft Translator has various feedback functions, such as B. the Collaborative Translations Framework, integrated into its products so that users can suggest alternative translations. These alternative translations will be built into the Microsoft Translator algorithms to improve future translations.

In November 2016, Microsoft Translator introduced deep neural network translation in many of the busiest languages. Neural networks provide better translation than industry standard statistical machine learning.

Web links

Individual evidence

  1. Microsoft Translator - Text Translation | Microsoft Azure Marketplace. February 6, 2018, accessed October 6, 2019 .
  2. ^ Languages. Retrieved October 6, 2019 (American English).
  3. ^ Arul Menezes at Microsoft Research. In: Microsoft Research. Retrieved October 8, 2019 (American English).
  4. ^ Neural Machine Translation. In: Microsoft Research. Retrieved October 8, 2019 (American English).
  5. Microsoft Translator: Microsoft Translator brings end-to-end speech translation to everyone with the world's first Speech Translation API. March 30, 2016, Retrieved October 8, 2019 (American English).
  6. Translator Text API - Microsoft Translator for Business. September 2, 2019, accessed October 8, 2019 .
  7. Microsoft Translator in the test: The online interpreter works that well. Retrieved October 8, 2019 .
  8. Microsoft Translator Hub: Discussion of BLEU Score. In: Microsoft Research. Retrieved October 8, 2019 (American English).
  9. Microsoft Translator: Microsoft Translator launching Neural Network based translations for all its speech languages. November 16, 2016, Retrieved October 8, 2019 (American English).