DeepL Machine Translation Provider for RWS Language Cloud
By Trados AppStore Team
Free
Description
This add-on enables users working in the cloud (through Trados Studio, Trados Team or Trados Enterprise) to receive machine translation results from the DeepL Translator - https://www.deepl.com/translator.
You can directly install the DeepL Machine Translation Provider from within your cloud account. Simply select your account icon in the top right hand corner of the screen> RWS AppStore.
Please note: You will need to purchase a subscription to DeepL Pro and you'll need an API key to use the add-on which you can obtain through DeepL sales. Pricing for the DeepL service is available through this DeepL website.
Technical details
1.5.6.0 - Trados Studio cloud capabilities, Trados Accelerate, Trados Team, Trados Enterprise
Changelog:
- Upgraded to app
- Added multi-region support
Checksum:
Release date:
Support website: https://community.rws.com/product-groups/trados-portfolio/rws-appstore/f/rws-appstore
Shared products:
Trados Studio cloud capabilities, Trados Accelerate, Trados Team, Trados Enterprise
1.5.6.0
- Upgraded to app
- Added multi-region support
1.4.13.0
- Fixed all chinese mappings
1.4.8.0
- fixed chinese simplified/traditional recognition issue
1.4.7.0
- Added more formality options ("prefer_less", "prefer_more")
- Changed contact email
1.4.5.0
- Add more details to API versions in dropdown
1.4.4.0
- Added possibility to choose API version
- Use dropdowns for formality and API version
1.3.6.0
- Glossary Support
1.1.2.0
- Disabled authorize attribute on AddOn controller
- Updated the hint that is display to user when selecting the formality setting
Language Cloud
Introduction
The Language Cloud add-on is available by "downloading" the add-on from here. When you "download" the application you will be redirected and asked to sign in to your language cloud account. If you don't have one, you won't be able to access the application at all.
Technical Requirements
- access to LC account
Key Features
There is no installation on your computer required as the installation takes place in your Language Cloud account where you will see something like this:
Clicking on the link will take you to the installation page where you simply click Install:
You'll then be asked to update needed configuration:
API Version
There are 2 choices:
- V1 - DeepL Advanced or Ultimate subscription
- V2 - DeepL Pro API subscription
API Key
DeepL API Key for selected API Version value. Example: xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx
Preserve Formatting
The Preserve Formatting setting controls whether the translation engine should retain the original formatting of the text, even when it would typically apply corrections.
Possible Values
Checked ✅ – The original formatting is preserved as closely as possible.
Unchecked ⬜ (Default) – The translation engine may adjust formatting for correctness.
Formatting Aspects Affected
Punctuation at the beginning and end of a sentence (e.g., extra spaces, missing periods).
Upper/lower case at the beginning of a sentence (e.g., maintaining capitalization exactly as in the source text).
Tag Handling
The Tag Handling setting specifies how the translation engine processes XML or HTML tags within the text. When enabled, the translation engine correctly processes and preserves structured content.
Available Options
• XML – Enables XML tag handling.
• HTML – Enables HTML tag handling.
• NONE – No specific tag handling applied.
Tag Handling JSON
The Tag Handling JSON field enables users to define tag handling settings for specific languages, overriding the global tag handling value.
Configuration Format
[
{"TargetLanguage": "ko-KR,de-DE", "TagHandling": "xml"},
{"TargetLanguage": "ja-JP", "TagHandling": "html"},
{"TargetLanguage": "fr-FR", "TagHandling": "none"}
]
Explanation
• TargetLanguage: Specifies one or multiple target languages using language codes.
• TagHandling: Defines how tags should be processed for the specified languages.
Supported values:
o "XML" – Handles tags as XML.
o "HTML" – Handles tags as HTML.
o "NONE" – No specific tag handling applied.
This feature enhances flexibility in managing language-specific tag processing.
Validation Rules
• The TagHandlingJson must be a JSON array.
• Each entry must contain:
o A valid language code in the format xx-XX (e.g., "en-US", "fr-FR").
o A TagHandling value that must be either "XML", "HTML", or "NONE".
• Invalid Configurations:
o Entries missing TargetLanguage or TagHandling.
o Invalid language formats (e.g., "English" instead of "en-US").
o Duplicate language entries.
o Incorrect TagHandling values (e.g., "text" instead of "HTML").
Ignore The Following Tags
This setting allows users to specify a comma-separated list of XML tags whose content should not be translated. This ensures that elements such as trademarks, product names, or technical terms remain unchanged.
Example Usage
If Ignore The Following Tags is set to keep, the text inside tags will not be translated.
Example Request:
Please open the page Settings to configure your system.
Example Response:
Bitte öffnen Sie die Seite Settings, um Ihr System zu konfigurieren.
Here, "Settings" remains in English because it is enclosed in the tag.
Translation Formality
Choose if the translated text should lean towards formal or informal language.
Possible options are: Default, More, Less, Prefer Less, Prefer More
Glossary Settings
This is a Language Pair to Glossary mapping: Enter your language pair to glossary associations in JSON format.
There are 2 ways in which you can map LPs to glossaries:
• Language Code:
{ "en>de":"glossary1", "de>fr":"glossary2"}
• Not detailed:
["glossary1","glossary2","glossary3"]
Split Sentences
The Split Sentences option controls how the translation engine divides the input text into sentences before translating. This setting is essential to ensure accurate sentence segmentation, especially when working with structured or complex content like HTML or XML.
Available Options
| Option | Description |
|---|---|
| 0 | No splitting: The entire input is treated as a single sentence. Recommended if your input text is already segmented. |
| 1 | Split on punctuation and newlines (default when Tag Handling is not set to html): Splits on punctuation and line breaks. |
| nonewlines | Split on punctuation only, ignoring newlines (default when Tag Handling is set to html): Ideal for HTML content where line breaks should be ignored. |
Default Behavior (If No Option Is Selected)
If you leave the Split Sentences field empty (unselected), the translation engine applies its default behavior, based on the Tag Handling configuration:
• If Tag Handling is not set to html:
o The default is 1.
o Sentences are split on punctuation and newlines.
• If Tag Handling is set to html:
o The default is nonewlines.
o Sentences are split on punctuation only, and newlines are ignored.
Recommendations
• Use 0 when your content is pre-segmented into sentences.
• Use nonewlines for structured documents like HTML, where you want to preserve line breaks without affecting sentence splitting.
• Leave the field blank for automatic default behavior, based on your Tag Handling settings.
Troubleshooting
Sometimes errors can be reported when trying to get a translation result from DeepL. This may be related to firewall restrictions which can be resolved by whitelisting the following domains:
www2.deepl.com
api.deepl.com
For more details on this you may find this link helpful:
https://gateway.sdl.com/apex/communityknowledge?articleName=000013232
