ContextuAIlize: Putting the
Sense in Context

A

ny localization professional knows the importance of context. If you don’t have a clear understanding of a text’s contextual underpinnings, then accurately communicating its meaning in another language is next to impossible.

As language technologies have grown more sophisticated, developers have spent untold hours introducing context tools into their localization products. At the 18th Process Innovation Challenge (PIC), which took place during June’s LocWorld53 conference, the importance of context was proven yet again when Agustín Da Fieno Delucchi took second place for his presentation, “ContextuAIlize: Automated Context for Effortless Localization.”

Access to that contextual information is often less than ideal. In fact, it can feel like a guessing game, given that it’s often missing from the string — and not a fun game with cool prizes, either. On the contrary, it collectively costs stakeholders millions or even billions of dollars.

“Every year, the software industry burns over $12 billion, just on low-value developer comments and back-and-forth clarification,” Agustín said in his presentation. “Meanwhile, the localization industry throws away over $100 million on avoidable rework.”

With fortunes like that on the line, it’s no wonder that ContextuAIlize struck a chord with PIC judges. So how does it work? As you might guess from the name stylization, it’s all about artificial intelligence (AI) — more specifically, an AI plugin that reads the project code and generates contextual information for the translator complete with comments, screenshots, and metadata.

“Instead of a blank comment field, translators get something like what you see in the XLIFF,” Agustín said. “Just enough to get it right the first time.”

That additional context can often mean the difference between first-effort success and frustrating time wasted on corrections. And that translates — no pun intended — into real-world time and cost savings, freeing up all concerned parties for more productive concerns.

“This isn’t just a translator user-experience upgrade,” said Agustín. “Post-editing time drops by more than 55 percent. Comment quality improves by 65 percent. Machine translation gets better, faster.”

Agustín put a fine point on the innovation by juxtaposing two images of a translation project. One portrayed a translator scrambling for more information due to insufficient context. The other showed the linguist moving on to the next task, confident in their work.

“Left: today’s reality — a vague string, a Slack message asking for clarification. Right: the ContextuAIlize experience — a clear comment, visual aid, and a generative AI suggestion that actually works. It’s a different world.”

That’s a world most translators seem happy to live in. If context is king in localization, then ContextuAIlize aims to have linguists feeling like royalty in no time.