SYSTRAN emphasizes human expertise by giving language experts a central role in the making of our professional translation model catalog. Access SYSTRAN state of the art neural translation technology through SYSTRAN Model Studio and become a translation model trainer!
«Generic» translation is available on all online tools and offers a quality that fits well in a general context. However it does not entirely meet the needs of professionals who are looking for a translation that is specific to their industry and business jargon.This translation customization is now facilitated by the learning capacity of neural translation engines. But this learning requires human expertise: training a neural engine is not a simple task. It requires a mastery of language and field of work. With In-domain translation models, SYSTRAN Marketplace aims to propose Financial, Medical, Legal or Automotive translation models, and so much more, in any language.Want to contribute and build your own translation models?
Start from your Translation Memory and upload your bilingual or monolingual in-domain corpus: a Spanish-English legal corpus for example and make it ready for model training. Your data remains entirely secured during the entire training process and not used for other purposes than your own model training.
Building a translation model from scratch is an arduous task. Select within SYSTRAN’s large translation model catalog the model that you want to bring to the next level and start feeding it with your domain specific data. By specializing an already trained model, you will benefit from embedded UD Sampling, Augmentation, Filtering, Noising and Tokenization.
Evaluate your model evolution at each training iteration with SYSTRAN model scoring module. With SYSTRAN model studio, compare BLEU score evolution of your models on more than 50 test files selected by SYSTRAN and categorized by domains. You can also add your own test set to check model’s progress on your very specific domain.
Once your in-domain model is ready, publish it to the SYSTRAN translation model catalog. Our worldwide community of professional end-users will be able to test your model and purchase it.You can also choose to publish your model privately if you aim to build it for a client. Intellectual property of the model remains yours.
AI behind translation model training needs significant computing resources (GPUs) to operate. Because infrastructure shouldn’t be your area of focus, we are providing you with the right computing resources to give you the best technical environment.And because we care about the environment, we chose to partner with OVH, European leader of cloud solutions and «green» hosting provider.
You possess multilingual data and are interested in monetizing them? You want to become a translation model trainer and sell them on our platform?
Fill-up this contact form and join the community!
Todos los campos son obligatorios
Alguna información no está debidamente cumplimentada
