Choosing What Matters Most to You | David Singleton | Google Zeitgeist

Choosing What Matters Most to You | David Singleton | Google Zeitgeist


My team, I’m personally based right here in
London, is responsible for Android Wear which is our platform for wearables based on Google’s
compute platform. And we’re really working very hard to bring
all of that power to you wherever you are, because a wearable product is fundamentally
always right there. You can glance at it and see useful information
and you can use it to get things done as you move around the world. Now, as we standard to decide what we wanted
to do in this space, we thought really deeply and felt very strongly that if you look at
what people wear, we all wear different clothes. We all carry different accessories. We all have a really individual sense of style. And, therefore, we didn’t think that a one-size-fits-all
model, where we would build a product ourselves, would actually work for wearables. So we set out to do everything I’m about to
show you in partnership with partners in this space that exist already, some of the electronics
companies that we were working with for phones and tablets, but also a range of new partners
from a watch industry. People like Tag Heuer or Michael Kors or Nixon. And we really fundamentally believe that we
can’t make these experiences accessible to everyone out there in the world without those
really important partnerships. So it’s exciting to be able to do this in
collaboration with many people around the world. And against that backdrop and what Behshad
was showing you is we do feel there’s tremendous potential to actually assist people with this
product that we’re building. So let’s take a closer look. If we switch to the demo here, you can see
my watch. And what you’re seeing here is the LG Watch
Sport which is a product from just one of our partners. And all of the Assistant use cases that Behshad
just demoed work here as well. So one of the things I like to do with this
watch is take it for a run. And these days, when I get home from the office,
I want to make sure I’m running in daylight. And, you know, it’s the time of a year when
that’s been changing so I can just go ahead and ask it questions like, “What time is sunset?” And get those answers right here. So lots of times, I go out directly for a
run. And everything that Behshad has already showed
you works right here on your wrist. We’re also using the Assistant to actually
control the things that you can do. Imagine if I was carrying a heavy bag of shopping
or imagine I’m in the kitchen and I’m cooking and I want to set the timer, I can simply
say, “Set a timer for ten minutes.” And while my hands have been busy, my watch
is keeping track of that time. And of course as I continue cooking I can
just glance right down and see how much time is left on that timer. So it’s great to have those things available,
but some of the places where this is most useful is actually where the Assistant works
together across devices to really help me. So I have a Google Home in my kitchen, and
I love to cook, and I use it a lot for doing things like setting timers or following recipes. But I don’t know about you, when I’m cooking
there are a few staple ingredients that I keep in my cupboards, like olive oil and so
forth, and I always find that when I run out of one of those things, I forget to replenish
it when I go shopping. So now what I do is I say to my Google Home
— you won’t see my watch respond here — I say, “Hey, Google, add soy sauce to my shopping
list.” And I’ve actually told the Google Assistant
where my local — where my local supermarket is so when I walk past or walk into the store,
I automatically get a card just like this that appears, and it reminds me when I walk
into the local store that I told it that soy sauce is what I need to replenish. And now I can cook a great Chinese dish the
next time I’m in the kitchen and not have a lack of that important ingredient. So as we’ve refined Android Wear, what we
find is that there are common pieces of information that people were asking for from the Assistant
on the watch all the time. Things like the weather or, I just showed
you, the time of sunset. And we’ve actually worked hard to make any
watch face, no matter what its style is, to show people those kind of information. So you can see right here I’m looking at the
weather, which is something we see a lot of questions for. But I can go ahead and long press on the watch
face here and decide that maybe instead of seeing the battery, I would like to see the
time of sunrise and sunset. And now when I go home, I’ll be able to see
if I’ve got time for that run without even having to ask a question. And, in fact, this is something we’ve heard
a lot. People have pieces of information that they
ask for habitually, and they actually vary a ton between people. For instance I travel a lot between the States
and the U.K., and one of the things I like to have on my watch face is the exchange rate
for those currencies, whereas for many other people that’s not of particular interest. So the ability to choose the information that
matters to you and have it available glanceably throughout the day is really, really valuable. And in fact, that’s kind of what watches are
for in the first place. The reason we have time on our wrist is to
help us make decisions as we go around our days. I want to be able to look down and say maybe
we’re getting towards the end of this meeting so I need to start bringing proceedings to
a close, or do I have time to grab a coffee before my next meeting. So we really see bringing these pithy answers
directly to your watch face as an evolution of what watches have always done for us. But these products now embed not just screens
and time keeping but also sensors. And probably what gets me most excited working
with wearables, if we could switch back to the slides, is their potential to help all
of us lead healthier and more active lives. So all of these Android Wear watches embed
the kind of features that you would expect from a fitness tracker. You can set goals. You’ll get notifications that help you stay
motivated to maybe work out a bit more through the day if you haven’t been so active. And they can also help you track workout sessions
actively. So if you were doing the spinning with Bradley
Wiggins, you’d be able to see just how hard you were working right on your wrist using
some of the sensors. But with a screen and the ability to apply
machine learning to the data coming from those sensors, we’re actually able to provide experiences
that go beyond just helping you track what you’re doing and actually show you how to
work out. So what you’re seeing in this video that’s
about to start playing behind me is a feature we built called Push-Up Challenges. And if you have never done a push-up before,
we can use that big screen to teach you how to actually go about doing the exercise, how
to keep the correct form. And then when you actually start doing those
press-ups — which, I’m sorry, I’m not going to do right now — we can actually analyze
the sensor data to see exactly how many you have done to help keep you using the correct
form so that you’re safe and having the most effective exercise. Now, that’s just one example of how we use
some of the sensors in these devices, but almost all smart watches these days are starting
to include a heart-rate sensor. And if we switch back to the demo here, I’d
love to show you something I’m pretty excited about that we can do with this in the future. So this is the typical trace that you would
see from — from the heart-rate sensor on the back of this watch. And this is a trace that I’m showing you that’s
come just from a piece of consumer hardware. In fact, it’s the watch that I’m wearing on
my wrist right now. And what these little peaks and troughs correspond
to is the blood flow that’s measured in my wrist by shining a light in and seeing how
much light comes back. And most products today take those peaks and
say look at how many of them there have been in a minute and give you your heart rate and
beats per minute. And of course that’s very interesting. But if we take a closer look, there’s actually
a ton more information right here. So if I draw a line straight through all these
peaks, that actually corresponds to respiration, how often I’m breathing. And of course that can tell you a lot about
how — how well or unwell I might be. If we look at the difference between the time
between these peaks, that gives you a measure of how variable my heart rate is. Heart rate variability, which is a leading
sign of health in general. And finally this little notch that you can
see here is called the dicrotic notch and the time between the peak and that dicrotic
notch shows you how fast the blood is moving through my veins, and that is something that
can be directly correlated with heart — with blood pressure. So let’s take a look at this. We have, with this — this is a Huawei watch,
and this is a little demo that my team have built. What you’re seeing at the top is the signal
coming from that blood flow in my wrist. You can see it’s detecting where there are
peaks and troughs. And what’s actually happening here is it’s
also compensating for movement. And eventually, when it gathers enough data,
it will actually tell you not only what my heart rate is, which is 100 beats per minute. I guess I’m a little bit stressed right now. [ Laughter ]
But also how variable it is. And, in fact, that’s not great. And at the top —
[ Laughter ] I was a lot more relaxed in rehearsals. [ Laughter ]
And at the top you’re also seeing the results of a machine learning model that is looking
for — and we can go back to the slides — is looking for and detecting arrhythmia in my
heartbeat. So it’s amazing what we can do combining machine
learning with actual consumer commodity hardware. And we’re partnering closely both inside Google
with the Verily team — we have a booth just outside and you can go and see more demos
there later — and also with institutions around the world, within the NHS and Manchester
in London, to be able to actually detect upcoming health conditions using these watches. And when we can catch conditions early, they
can be treated much more effectively. When we can give people that kind of coaching
as they live their lives to help them stay with a course of prescribed exercise or medication
or whatever, it makes their treatment much more effective and much more cost effective,
and so I think there’s tremendous potential here for the future and we’re very excited
to be working on it partnership with all of those folks I talked about at the beginning. Thank you.

Leave a Reply

Your email address will not be published. Required fields are marked *