With our latest update to the PCS Communication engine, PCS Spark and ALEX can now understand you even better thanks to our enhanced contextual understanding.
Dealing with context is one of the more difficult things a Speech AI will need to do throughout a conversation. In previous versions of the PCS Engine, each question was handled independently without consideration of what had been said previously.
With these latest improvements however, the PCS Engine takes into account what had been discussed previously and its relevance to the current question. Why does this matter you might wonder?
Let’s look at an example of how the PCS Engine deals with pronouns. During an interaction with a learner the topic of family history might come up. A learner might ask about parents, siblings, etc.
When a learner asks “Is your Dad still alive?” The digital PCS Spark patient may respond with “Yes.” A natural follow up to that question then is “Is he in good health?” In this example, the digital patient has both a Dad and a Brother both of which would be considered “He”. With the latest updates to our PCS Engine, we are now able to associate “He” with “Dad” and provide the correct response to whether or not he is in good health based on the wording of the previous question.
Adding even a basic form of contextual understanding is so challenging that our engineers are now training dedicated natural language processing neural networks solely for identifying the words and concepts that follow-up questions may refer to with various pronouns.
All said, this update will provide learners with a more natural interaction with their virtual patients using PCS Spark. We are committed to continuing to improve the PCS Communication Engine so that it remains at the forefront of AI and Machine Learning in the Healthcare Simulation Space.
If you’re not a PCS Spark user yet, sign up to see a first hand demo of this powerful software.
