Azure-Translator-Document-translation

Version: 1

Microsoft•Last updated December 2025

Document translation is a cloud-based, multilingual service that uses AI to translate documents from one language to another while preserving the document layout.

Azure Translator

Azure Translator is part of the Azure AI Foundry tools family and powers many Microsoft products and services used by thousands of businesses worldwide for language translation operations. Azure Translator supports features such as text translation, document translation and custom translator and supports 135 languages .

Key capabilities

About this model

Azure Translator's Document translation translates complex documents while maintaining the document fidelity.

Key model capabilities

File type support

Supports various document formats including .docx, .ppt & .txt. Full list

Secure & Compliant

Enterprise grade security with data privacy & compliance.

Flexible integration

Use REST APIs, SDKs, connectors or containers.

Customize

Build custom models to fine tune quality.

Use cases

Outbound Translation: Translate documents for publishing in multiple languages (e.g., manuals, reports, marketing collateral). Machine translation works best for structured, less creative content.
Inbound Translation: Translate incoming foreign-language documents for internal understanding or automated processing (e.g., business reports, compliance docs, e-Discovery).
Bidirectional Collaboration: Enable multilingual document exchange in workflows or enterprise systems.
Post-edited Translation: Machine translation followed by human review for high-quality output. Recommended for critical or customer-facing documents.
Raw Translation: Direct machine output without human review for low-risk scenarios. Lowest cost but higher error tolerance.

Key use cases

Enterprise workflows, apps, and portals requiring multilingual document support.
Automated document pipelines for compliance, analytics, or content distribution.

Out of scope use cases

Sensitive or High-Stakes Scenarios: Include human review for medical, legal, or regulatory documents.
Awarding or Denying Benefits: Not designed for decisions involving insurance, loans, or rights.
Non-professional Content: Colloquial, social media, or transcribed speech may yield poor results.
Legal & Regulatory Compliance: Ensure adherence to applicable laws and codes of conduct.

Pricing

Pricing is based on characters. Pricing page

Technical specs

Document translation is part of Azure Translator model family supporting 135 languages.

Training cut-off date

The model training process is iterative and does not follow a fixed cadence.

Input formats

Word, Excel, Outlook, PowerPoint, CSV, HTML, MHTML, Plain Text, TSV, XLIFF.

Supported language

130+ languages are supported including Arabic, Bengali (Bangla), Chinese Simplified, Chinese Traditional, Dutch, English, French, French (Canada), German, Hindi, Italian, Japanese, Korean, Malay, Portuguese (Brazil), Portuguese (Portugal), Punjabi, Russian, Spanish, Tamil, Telugu, Thai, Turkish, Urdu, Vietnamese. Full language support list: Azure Translator language support API reference for supported language list: Reference

Supported Azure regions

See the full list of supported Azure regions for Azure Translator linked here .

Sample request

curl --request POST \
  --url 'https://{your-document-translation-endpoint}/translator/document:translate?api-version=2024-05-01&sourceLanguage=en&targetLanguage=fr' \
  --header 'Content-Type: multipart/form-data' \
  --header 'Ocp-Apim-Subscription-Key: <your-subscription-key>'
  --form 'document=@<path-to-your-document>/your-document-file.docx' \
  --output translated-document-fr.docx

On success, the HTTP response body is the translated document (binary) — the same file format as the source.

Model architecture

Transformer-based multilingual architecture optimized for high-quality, context-aware neural machine translation. Various libraries, services and packages for document layout retention.

Long context

Information not available.

Optimizing model performance

The batch method is suitable for most use cases and offers additional file type support (PDF included) and the best quality. Use the synchronous method only for real time scenarios.