Q And how quickly does that work? How much do you need to train?
A So, uh, the, the good thing is, it's, these are, sort of, uh, don't require any, so in the case of signal processing, it doesn't require any training, but, um, if only, well, it does require training of, of the user, so we, we provide them with, like, there, it's a class of two or three parameters, and, um, so we've essentially tuned the, tuned them, um, Um, with, so that the defaults mostly work. Um, the, the difficulty with, um, something I didn't really touch on with, with this problem. So imagine we're talking about monitoring, which means that when we trigger an alert, we ring the phone, whatever the time of day or night it is. So any false positives, our, our users are like, ah, um, luckily it doesn't happen too often, but it's just sort of a, a high, highly emotional response to our product. Um, Especially from the part of the spouses when it's three a.m. So, so we spent, and it's really the hard part about the, the science part of it is the quality is a really difficult problem. So even if we do a, um, which I like to think we're getting close to, but if we do a perfect job, you're just meeting your customer's expectations. So it's a, there's a lot of pressure behind that. That's why we invest a lot of time and money to try to figure out, well, how do we crack that?
AI assessment note: “in the case of signal processing, it doesn't require any training”