First introduced earlier this year, AIQE provides quality estimation scores for texts that are translated through machine translation engines in memoQ TMS. These quality estimates are made available via integrations with ModelFront and TAUS. This innovative technology addresses the risks associated with the inconsistent quality of machine translation, aiming to reduce the time, effort, and resources required to ensure machine-translated content’s readiness and accuracy.
New AIQE functionality enables less time spent on post-editing
With memoQ 10.4, project managers can reach an unapparelled level of automation and workflow optimization when dealing with Machine Translation Post-Editing (MTPE). The AI-based technology now offers the opportunity to set up condition-based workflows with the possibility of defining a threshold at which machine-translated segments are automatically validated and where post-editing can be skipped. This ensures that time and money spent on post-editing focuses on segments that need particular attention.
Auto selection of MT engines in pre-translate with AIQE
memoQ’s latest version also introduces new automation opportunities for customers who rely on multiple machine translation engines in their localization workflow. No need to worry about which MT providers will perform best for specific projects and segments. AIQE assesses which MT engines provide the best quality scores for each segment and automatically performs pre-translation accordingly!
Evaluating the performance of machine translation engines in projects
Another step towards increased efficiency when working with machine translated content; match rates now appear in the .MQXLIFF file associated with a given project. This enables the user to see which MT engines the matches come from, and which one performed better in the given document. This feature is especially useful when evaluating machine translation engines, which can ultimately help the user make more informed decisions for future projects.