Infor needed a reliable machine translation (MT) solution to improve quality, maintain competitive translation project turnaround times, and expand their language offerings to meet global demands.
Managing high-volume translations, keeping consistency across multiple languages, reducing manual tasks, and maintaining quality under tight deadlines.
Infor’s statistical machine translation (SMT) struggled with unrelated language pairs (like English to Japanese) and disfluent translations. Limited training data caused terminology issues when combining datasets, making it hard to maintain consistency across projects.
“As Infor’s preferred machine translation provider, globalese by memoQ has been a reliable and strong partner for almost a decade now, allowing our department to speed up the translation process of new and recurring — often large — projects while maintaining structural, terminological, and overall linguistic correctness.”
John Musters
Technical Translation Support Specialist Infor
Infor had previously relied on SMT, which worked well for European language pairs such as English to German but struggled with unrelated pairs like English to Japanese. This limitation confined their machine translation efforts to a narrow scope and modest range of target languages. globalese by memoQ changed this by introducing neural machine translation (NMT), enabling Infor to expand their translation projects globally. They now confidently translate into additional languages, including those languages mentioned above, addressing diverse market needs.
While SMT provided generally usable translations, it often lacked fluency, making the results feel unnatural. By switching to NMT with globalese by memoQ, Infor achieved a significant quality boost.
Translations are now not only more accurate but also more fluid and natural, enhancing the user experience across their multilingual products. An additional benefit of globalese by memoQ is its ability to correctly maintain the structural integrity of the individual segments in the source files, often in DITA XML, or JSON format.
One of Infor’s primary challenges was the limited availability of high-quality training data for certain projects. Combining data from various sources often led to terminology inconsistencies, making it difficult to ensure accurate translations. globalese by memoQ’s in-domain adaptation functionality addressed this issue by differentiating between core and auxiliary corpora. This enabled Infor to translate even small-scale or new projects effectively, maintaining consistency across all outputs.
Additionally, Infor’s decision to run globalese by memoQ as an in-house system has reinforced data security and optimized internal workflows. This setup not only ensures confidentiality but also reduces the need for extensive security checks when connecting to other systems.
Infor is a global company that provides enterprise software products for every aspect of business. We build complete industry suites in the cloud and efficiently deploy technology that puts the user experience first, leverages data science, and integrates easily with your existing systems. We have 125+ products of which many have a translated User Interface or translated Help documentation. Depending on the project, we translate up to 40 different languages. In total, about 20 million words are translated every year.