Translator On-Demand Glossary Translation Service

Overview

This reusable service gives every app and flow one governed, on-brand translation endpoint built on Microsoft Translator V3. It is triggered over HTTP (a Request trigger, callable from a Power App, another flow, or any client). It applies a company glossary stored in Dataverse as Microsoft Translator dynamic-dictionary markup so brand/product terms always translate the approved way, returns the translated text, and writes a usage/audit record to Dataverse for cost tracking. Without a shared service, each app re-implements translation inconsistently and brand terms drift; a central endpoint enforces the glossary, standardizes behavior, returns a correlation id for tracing, and tracks spend (character counts) by request.

Use Case

Multiple apps and flows need translation. Rather than each one calling Translator directly (and re-defining glossary terms, response handling, and logging), they POST { text, to, from } to this single endpoint and get a consistent, glossary-aware result plus centralized usage logging. Typical callers: a Power App translate button, a support-ticket flow localizing replies, a marketing flow localizing copy.

Flow Architecture

Environment Variables

Connectors & Connections

Connector	API name	Actions used
Microsoft Translator	shared_microsofttranslatorv	Translate
Microsoft Dataverse	shared_commondataserviceforapps	ListRecords CreateRecord
RRequest	request	Request Response

Note — All connections are referenced as solution connection references; the flow is portable between environments as long as a connection is mapped at import time.

Customization Guide

Almost every realistic variant of this flow can be implemented by changing environment variable values. A few cases require small edits inside the flow definition — those are called out explicitly below.

Glossary

Add rows to the Translation Glossary table (Source Term, Translated Term, Target Language). Only entries matching the resolved target language are applied.

Default language

Change flowlibs_DefaultTargetLanguage to control the fallback when callers omit to.

Endpoint security

The Request trigger is unauthenticated by default; add a shared-secret header check (Condition on x-api-key) or front it with a Power App that holds the secret.

Document mode

For full files, switch to Translator V3 Document Translation (StartDocumentTranslation / GetTranslationStatus), which is async.

Caching / cost

Cache common phrase translations in Dataverse keyed by (text, target) to cut Translator spend; the usage log already captures character counts for attribution.

Detected language

To store the auto-detected source code, add a built-in HTTP call to the Azure AI Translator /detect endpoint (endpoint/key/region from env vars).

Key Expressions

The flow is intentionally light on Power Fx / WDL gymnastics — the heaviest expressions are the branch-name concatenation and the approval outcome check. They are listed below in the order they appear in the flow.

@coalesce(triggerBody()?['to'], parameters('EnvVar_DefaultTargetLanguage'))

@concat('flowlibs_targetlanguage eq ''', variables('varTargetLanguage'), '''')

@replace(variables('varProcessedText'), items('Apply_Glossary_Terms')?['flowlibs_name'], concat('<mstrans:dictionary translation="', items('Apply_Glossary_Terms')?['flowlibs_translatedterm'], '">', items('Apply_Glossary_Terms')?['flowlibs_name'], '</mstrans:dictionary>'))

@first(body('Translate_Text'))?['TranslatedText']

@length(triggerBody()?['text'])

Customize & download