Co-op Translator builds a new paradigm for collaborative translation

By /

Microsoft has released Co-op Translator, an open source tool designed to simplify the process of multilingual translation, especially for technical documents and images embedded in text. It combines Azure OpenAI and Azure Computer Vision services to efficiently complete the translation of Markdown files and images, helping developers break through language barriers and promote projects around the world.
The translation of 276 images and 153 Markdown files could be completed in less than 2 hours.

1. Project Introduction

GitHub project address: Azure/co-op-translator

PolyglotSync is aWeb-based collaborative translation platform, the main features include:

  • Support the allocation and management of multilingual translation tasks
  • Integrate with Azure Translator (which can be replaced with other services) as a machine for initial translation
  • Support manual proofreading, terminology unification and context management
  • Support user rights management and collaboration process control

2. Architecture and core components

The project is mainly divided into the following parts:

  1. Frontend

    • Build a modern UI with React
    • Provides functions such as translation text comparison, term highlighting, and context viewing

  2. Backend

    • Provide RESTful APIs with Node.js + Express
    • Manage translation tasks, glossary, user rights, etc.

  3. Translation Service

    • Azure Translator API is used by default
    • Supports automatic translation of first drafts and can be replaced with translation engines such as DeepL and Google as needed

  4. Database

    • Use MongoDB to store translation tasks, user information, and terminology data

3. Core advantages

  1. Human-machine collaboration (Co-op Mode)

    • Machine translation is based and manual secondary optimization is performed
    • Changes can be tracked at every step

  2. Terminology consistency

    • Terminology can be uploaded, and the system will automatically highlight and provide suggestions during translation

  3. context-aware

    • Support paragraph level context viewing to avoid machine literal translation of “words that do not express their meaning”

  4. real-time collaboration

    • Multi-person online collaborative translation makes task splitting more efficient

4. Application scenarios

  • Localization of enterprise-level product documents
  • International website or SaaS platform
  • Construction of multilingual customer service knowledge base
  • Translation of multilingual course content in the education industry

5. Deployment method

The project supports local or cloud deployment:

git clone https://github.com/Azure/co-op-translator.git
cd co-op-translator
npm install
npm start

and .env Configure the API key for Azure Translator or other translation service in the file.

6. Summary

The emergence of the PolyglotSync project allows us to seeThe future of “translation is collaboration”。AI is no longer a tool to replace humans, but has become a useful assistant in improving human language. If you are processing and managing multilingual content, this project is undoubtedly a platform worth trying.

GitHub：https://github.com/Azure/co-op-translator

Oil tubing:

Scroll to Top