Our language catalog has been built on the best linguistic practices acquired over the past 50 years, combined with state of the art of neural technology and the new capacity it offers in terms of engine training. This combination of linguistic expertise and advanced neural technology is unique on the market and offers professionals the best translation quality using their tone of voice and terminology.
SYSTRAN Pure Neural Server offers unique customization features to allow your teams to continuously improve the translation quality.
Use your own assets: dictionaries, glossaries, translation memories with our Linguistic Asset Management tool. Receive and manage user feedback easily.
Benefit from a unique feature ‘Coding Engine’: when a Neural MT engine has been trained with this technology, users can then add their own User Dictionaries and their terminology will be instantaneously translated as requested.
Users don't need to be linguists or developers to create dictionaries. Our Coding Engine, a linguistic tool with 50 years of knowledge behind it, will do all the work for them so that all inflected forms of their terms will be translated.
Achieve greater accuracy with SYSTRAN domain-specific engines and specialized translation models. Translation models can be built based on your existing corpus (bilingual or monolingual text). If your company doesn't have enough data, we can provide access to our own Corpus Factory, a substantial internal database encompassing monolingual and bilingual text with free or purchased intellectual property rights.
Your data is gold! We can re-train neural network engines with your own data. A specialized engine outperforms any generic engine for in-domain translation.