An AI solution that doesn’t leave patient care in limbo

Have you ever tried dictating a text message to a friend or asking a voice assistant a question, only to find that your phone completely misunderstood you? Chances are, if you’ve used any major smartphone software in the last decade, you’ve done this. This is often because these systems cannot decode language in incomplete situations. And most of the time the real world isn’t perfect.

However, new applications of IBM’s Granite language model show that the model can understand people much more easily than other major language models, even in noisy and stressful scenarios. This has the potential to open up countless new use cases for LLMs where the primary input is speech.

An example can be found in the skies over Australia. The Royal Flying Doctor Service (RFDS) is a not-for-profit organization that works across the country to help people far from city centers access medical professionals and services. Australia’s largest population centers are located primarily around…