Skip to content

About this Course

MTUOC: A Practical Course on Machine Translation Technologies is an open online course designed to introduce learners to practical machine translation technologies through the MTUOC ecosystem.

Machine translation is now part of many professional translation, localisation and multilingual communication workflows. However, many open-source machine translation and large language model tools require technical knowledge that may be unfamiliar to translators, language professionals and students in the humanities. This course addresses that gap by offering a practical, guided and accessible introduction to machine translation technologies.

The course focuses on MTUOC, a modular ecosystem of open-source tools designed to make machine translation workflows easier to deploy, evaluate, integrate and adapt. Through a set of step-by-step activities, learners will explore how machine translation systems can be used, evaluated and connected to professional translation environments.

The course is organised into four modules:

  1. Deployment, Integration and Evaluation
    Learners will explore how to start machine translation systems, use MTUOC Server, connect MT systems to translation workflows and evaluate translation output with automatic metrics.

  2. Corpus Engineering
    This module introduces the preparation, cleaning, alignment and conversion of linguistic data for machine translation purposes.

  3. NMT Training and Fine-tuning
    Learners will work with neural machine translation workflows, including model training, adaptation and deployment.

  4. LLMs for Translation
    This module introduces the use of large language models in translation-related workflows, including prompt-based translation and fine-tuning approaches.

The course is aimed at translation students, language professionals, teachers, researchers and anyone interested in understanding how machine translation technologies work in practice. No advanced programming background is required, although learners should be willing to work with digital tools, follow technical instructions and experiment with command-line environments when needed.

By the end of the course, learners will be able to:

  • understand the main components of a machine translation workflow;
  • deploy and test machine translation systems using MTUOC tools;
  • evaluate machine translation output with automatic metrics;
  • prepare and manage linguistic resources for machine translation;
  • understand the basic logic of neural machine translation training and fine-tuning;
  • explore the role of large language models in translation workflows;
  • critically reflect on the possibilities and limitations of machine translation technologies.

Support and acknowledgements

Organized with the support of the Faculty of Arts and Humanities of the Universitat Oberta de Catalunya (UOC). This open course is organized within the framework of the Master's degree in Translation and Technologies.

This work is also supported by the projects:

  • TamTAS: Terminology-Aware Machine Translation for Accessible Science, project PCI2025-167063-2, founded by MICIU/AEI/10.13039/501100011033 and by the European Union.
  • LLMTrad-IBE: Large Language Models for Translating Low-Resource Romance Languages of the Iberian Peninsula, project PID2024-158157OB-C33 founded by MICIU/AEI/10.13039/501100011033/ FEDER, UE.

Licence and reuse

This course is conceived as an open educational resource. Its materials are intended to be reusable, adaptable and extensible for teaching, self-learning and research training contexts.

Users may follow the full course sequence or reuse individual modules, tutorials and activities according to their own learning, teaching or training needs.