Text Translation

Supported Languages

Use Cases

Multilingual chatbots

The purpose of chatbots is to answer customer queries. Chatbots can detect the query language and answer the question in the same language.

Developers that create the bots might not speak or can program the chatbots in a dozen languages. Instead, developers can use Language to detect the query language and perform automatic translations to and from the chatbot's original language.

Online advertisements

Online advertising companies crawl millions of web pages to analyze the content surrounding the ads and ensuring brand safety. Brand safety helps advertisers host their ads in the right place for the right kind of audience. For example, brand safety suggests avoiding pizza ads in weight loss articles.

Typically, ad companies crawl web pages that are in regional languages. Before classifying the text, advertisers often translate the content to English. Language provides near real-time translation combined with text classification allowing advertising companies to create a single language classification model.

Document search

Companies operating in different industries have tons of documents such as request for proposals (RFPs), product documentation, services, and purchase agreements. These documents are often stored in document databases, which require keywords for indexing and to speed up search and retrieval.

To enable document search even when a query is in a different language, businesses can use Language to translate the query into the set core popular languages, whenever few results are in the original language.

Examples

Preventing Translation of Content

The translation engines allow you to tag content so that it isn't translated. For example, you might want to tag code, a brand name, or a word/phrase that doesn't make sense when localized. To prevent translation of content, complete the following:

1. Tag the content with no translate. For example:

   POST /api/translate/fr/en
   [  
       "En français, ils appellent le musée du Louvre <span class=\"notranslate\">le musée du Louvre</span>",
       "En français, ils appellent le musée du Louvre <span>le musée du Louvre</span>",
       "En français, ils appellent le musée du Louvre <div class=\"notranslate\">le musée du Louvre</div>",
       "En français, ils appellent le musée du Louvre <div>le musée du Louvre</div>"
   ]
    
   Result:
   [
       "In French, they call the Louvre Museum <span class=\"notranslate\">le musée du Louvre</span>",
       "In French they call the Louvre Museum <span>the Louvre Museum</span>",
       "In French, they call the Louvre Museum <div class=\"notranslate\">le musée du Louvre</div>",
       "In French they call the Louvre Museum <div>the Louvre Museum</div>"
   ]

2. Tag content with translate="no". For example:

   POST /api/translate/fr/en
   [  
       "En français, ils appellent le musée du Louvre <span class=\"notranslate\">le musée du Louvre</span>",
       "En français, ils appellent le musée du Louvre <span>le musée du Louvre</span>",
       "En français, ils appellent le musée du Louvre <div class=\"notranslate\">le musée du Louvre</div>",
       "En français, ils appellent le musée du Louvre <div>le musée du Louvre</div>"
   ]
    
   Result:
   [
       "In French, they call the Louvre Museum <span class=\"notranslate\">le musée du Louvre</span>",
       "In French they call the Louvre Museum <span>the Louvre Museum</span>",
       "In French, they call the Louvre Museum <div class=\"notranslate\">le musée du Louvre</div>",
       "In French they call the Louvre Museum <div>the Louvre Museum</div>"
   ]

Inline dictionary

If you already know the translation you want to apply to a word or a phrase, you can supply it as markup within the request.

The inline dictionary feature is safe only for compound nouns such as proper names and product names. We recommend you use this feature sparingly.

It's by design that the inline dictionary only works with format set to html.

Syntax:

<oratrans:dictionary translation="Bonjour">Hello</oratrans:dictionary>

For example:

POST /api/translate/en/fr
[
    "You can use <oratrans:dictionary translation=\"The Cool Text App\">cool text application</oratrans:dictionary> to create some cool text",
    "You can use cool text application to create some cool text"
]
 
Result:
[
    "Vous pouvez utiliser The Cool Text App pour créer du texte cool",
    "Vous pouvez utiliser l'application de texte cool pour créer du texte cool"
]

Known Issues and Limitations

Oracle Translate provides good quality and reliable translation for a wide range of business enterprise and generic content. Translated text can be utilized without further modification if some imperfections are tolerated. In cases where precision is essential, post-editing by native language speakers can be commissioned to rectify and enhance machine translation output.

To enhance the quality of your translations, consider these potential limitations and adopt the recommendations listed below: