Dynamics 365 Business Central: How to use Custom Translator (D365 Translation Service support ends)

Hi, Readers.
Today I would like to talk about how to use Custom Translator.

As you might, the Microsoft Dynamics 365 Translation Service (DTS) is hosted in Microsoft Dynamics Lifecycle Services (LCS). It’s designed to enhance the experience for partners and independent software vendors (ISVs) when they translate their solutions or add a new language for supported Dynamics products. We have briefly discussed this before:

However, Microsoft announced in the Dynamics 365 Translation Service overview that this service will reach end-of-support on October 15, 2025, and will no longer be available after this date.

We’re announcing the deprecation of the Dynamics 365 Translation Service (DTS). This service will reach end-of-support on October 15, 2025, and will no longer be available after this date. Learn more in Microsoft Dynamics 365 Translation Service support ends on October 15, 2025.

Microsoft Dynamics 365 Translation Service support ends on October 15, 2025:

In the document, Microsoft recommends other translation services, such as Microsoft Custom Translator. In this post, I will briefly test how to use it.

Custom Translator is a feature of the Azure AI Translator service, which enables enterprises, app developers, and language service providers to build customized neural machine translation (NMT) systems. The customized translation systems seamlessly integrate into existing applications, workflows, and websites.

Custom Translator:

PS: To use the Azure AI Custom Translator portal, you need the following resources:

A Microsoft account.
Azure subscription – Create one for free

Once you have an Azure subscription, create a Translator resource in the Azure portal to get your key and endpoint. After it deploys, select Go to resource.

PS: If you do not find an available subscription, please switch to the paid environment.

Subscription. Select one of your available Azure subscriptions.
Resource Group. You can create a new resource group or add your resource to a preexisting resource group that shares the same lifecycle, permissions, and policies.

Resource Region. Choose Global unless your business or application requires a specific region. If you’re planning on using the Document translation feature with managed identity authorization, choose a geographic region such as East US.

Name. Enter a name for your resource. The name you choose must be unique within Azure.

Pricing tier. Select a pricing tier that meets your needs:

Each subscription has a free tier.
The free tier has the same features and functionality as the paid plans and doesn’t expire.
Only one free tier resource is available per subscription.
Document translation is supported in paid tiers. The Language Studio only supports the S1 or D3 instance tiers. If you just want to try Document translation, select the Standard S1 instance tier.

Select Review + Create.

Review the service terms, and select Create to deploy your resource.

After your resource successfully deploys, select Go to resource.

You need the key and endpoint from the resource to connect your application to the Translator service. Paste your key and endpoint into the code later in the quickstart. You can find these values on the Azure portal Keys and Endpoint page:

All Azure AI services API requests require an endpoint URL and a read-only key for authentication.

Authentication keys. Your key is a unique string that is passed on every request to the Translation service. You can pass your key through a query-string parameter or by specifying it in the HTTP request header.
Endpoint URL. Use the Global endpoint in your API request unless you need a specific Azure region or custom endpoint. For more information, see Base URLs. The Global endpoint URL is api.cognitive.microsofttranslator.com.

These keys are used to access your Azure AI Foundry API. Do not share your keys. Store them securely– for example, using Azure Key Vault. We also recommend regenerating these keys regularly. Only one key is necessary to make an API call. When regenerating the first key, you can use the second key for continued access to the service.

For more information, see how to create a Translator resource.

Once you complete the prerequisites, sign in to the Azure AI Custom Translator portal to create workspaces, build projects, upload files, train models, and publish your custom solution.

After your sign-in to Azure AI Custom Translator, you’ll be asked for permission to read your profile from the Microsoft identity platform to request your user access token and refresh token. Both tokens are needed for authentication and to ensure that you aren’t signed out during your live session or while training your models.
Select Yes.

Select Create a new workspace.

Type Workspace name and select Next, for example, BC Translator.

Select “Japan East” for Select resource region from the dropdown list.

Region must match the region that was selected during the resource creation. You can use KEY 1 or KEY 2.

Select Next.

Done. Once the workspace is created successfully, you see the **Projects** page.

Select Create project.

Type Project name. And select English (en) as Source language from the dropdown list. Select German (de) as Target language from the dropdown list. Select General for Business from the dropdown list.
Then select Create project.

In order to create a custom model, you need to upload all or a combination of training, tuning, testing, and dictionary document types.

As a test, I directly uploaded the Microsoft translation file as a dictionary first. ~Dynamics.365.BC.36321.W1.DVD\applications\BaseApp\Source\German language (Germany).Source\Translations

The file is 79.3MB.

Select Add document set.

Check the Dictionary set box and select Next.

I used the Sentence dictionary in my test.

For XLF files, please select Translation memory (TM) file.

Select Upload file, then choose upload.

But because the XLF file was too large, I couldn’t finish uploading it even after an entire afternoon… Later I edited the file, reduced the content, and uploaded it successfully.

After clicking, you can confirm the content inside, which is automatically generated from the XLF file.

PS:
1. This translation file can be downloaded.

But it will be split into two files according to the language, which contains the translated content.

en:

de:

Now you’re ready to train your English-to-German model.

Select Train model from the left pane.

Type en-de with sample data for Model name.
Keep Dictionary-only training checked.

Under Select documents, check sample-English-German and review the training cost associated with the selected number of sentences.
Select Train now.

Select Train to confirm.

However, it failed in my test.

But it doesn’t matter, let’s continue testing.

PS: In Microsoft Quickstart: Build, publish, and translate with custom models, Microsoft show you how to upload training documents for customization.

Finally let’s look at how to use Translate text. There are two main ways

Developers should use the Category ID when making translation requests with Microsoft Translator Text API v3. More information about the Translator Text API can be found on the API Refe r ence webpage.

Here I use Postman to do the API test:

Endpoints:
Text Translation: https://api.cognitive.microsofttranslator.com/
Document Translation: https://je2.cognitiveservices.azure.com/

Params:

api-version=3.0
from=en
to=de
category=2a6b36f2-b888-40aa-89a3-965d8259e82 (Since the training failed, it is only for reference)

Headers:

Ocp-Apim-Subscription-Key=9qtViukd9YSG4IkJdRgZksGc7JiKyNBIYT7stU5Dr51RM8UOneWPJQQ
Ocp-Apim-Subscription-Region=japaneast
Content-Type=application/json

Body:

[
    {
        "Text": "My Incoming Documents"
    }
]

Done.

The category parameter is not required, but if you don’t fill it in, it will be translated mechanically instead of using the AI you trained.

In the test, I only translated one text, but you can also translate multiple texts at the same time. More details: Text API v3.

Using this API, although it is relatively complicated, it is possible to automatically translate the text in BC according to the user’s language. But as of now, I personally don’t recommend doing this…….

Business users can download and install our free DocumentTranslator app for Windows.

This is a Windows app that can directly translate text and files. I will test it next time I have the chance.

The Document Translation tool translates documents in your local storage using the Microsoft Azure Translator service. It can translate Office documents (Word, Excel, PowerPoint), HTML documents, PDF documents, Outlook Messages, Markdown, MHTML, plain text, RTF and XLIFF files.
DocumentTranslation | Command Line tool and Windows application for document translation, a local interface to the Azure Document Translation service for Windows, macOS and Linux.

Here is a complete guide:
DocumentTranslation | Command Line tool and Windows application for document translation, a local interface to the Azure Document Translation service for Windows, macOS and Linux.

Great, give it a try!!!😁

END

Hope this will help.

Thanks for reading.

ZHU