The right words in the right place, in the right way

eCommerce and other sectors with international audiences have shown how quickly businesses can spread into new geographies, often opening huge opportunities for growth. However, this also places a demand on the business to ensure that the content and material that drive customer experiences in these markets are made relevant and correct.

This has fuelled a huge interest in translation and localization, as 87% of non-English speaking customers would not buy from an English-only website. (CSA Research, 2020)

XTM joined the MACH Alliance in March 2023 to elevate localization technology within the MACH ecosystem.

Having worked with fellow Alliance members such as ContentStack, Kontent.ai, and Akeneo, XTM seeks to not only drive localization within composable architectures but also to explore the impact of new AI technologies.

As a recent example, XTM Cloud’s Auto-inline Placement Tool uses AI to automatically recognize and place inline tags in the right place. This drastically reduces the amount of manual work required, saving up to 200 hours per month for a company’s average translation volumes.

In the immediate short term, the use of AI will also enable beta testers to experiment with the automatic identification of offensive or discriminatory language within XTM Workbench. Furthermore, XTM Workbench will highlight and mark as errors sentences that fall below a specific quality standard.

Each segment, created by humans or AI, will be assessed based on grammar, typos, appropriateness, and accuracy.

Elsewhere, Neural Machine Learning—a key AI feature in language technology—is part of XTM’s Translation Memory (TM) that combines cutting-edge TM and Neural Machine Translation (NMT) functionality.

When a sentence is already partially translated, the NMT engine can use existing matches to fill in the gaps and deliver a complete translation, or even adjust the fuzzy match if needed. As neural machine engines train themselves ‘on the fly’, using translation memory in a self-sustaining, interactive process, this innovation paves the way for high-quality translation at scale.