How AI could help translate extreme weather alerts

A startup that provides AI-powered translation is working with the National Weather Service to improve language translations of extreme weather alerts across the U.S., Axios’ Ayurella Horn-Muller reports.

The big picture: Gaps in language access to emergency alerts during extreme weather events have led to missed evacuations, injuries and loss of life for non-English speakers. The NWS hopes machine learning could mitigate that.

How it works: Incorporating a mix of software and human translators, AI-translation service Lilt learns from linguists in real time using a neural network, or a computer system modeled loosely on the brain — which gets smarter with each use.

  • The software gets used by human forecasters at the NWS forecasting office, with the AI engine suggesting translation for the translators to work with, while actively storing their input.

What they’re saying: Phil Stiefel, solutions lead at Lilt, told Axios there is a longstanding need for better translation at the NWS.

  • “If there’s a translation error in a translated weather report, and somebody takes the wrong action based on that missed translation, then somebody could get hurt or even killed because of that,” Stiefel says.

Of note: A 2021 research article in the American Meteorological Society looked at issues in English-to-Spanish translation of weather alerts that didn’t account for dialects, which led to “inconsistent risk messaging,” in Spanish-language alerts.

Read the whole story.