Skip to main content

Let the robots speak to one another

Let the robots speak to one another


Voice is the ultimate wireless communication protocol

Share this story

Amazon Echo next to Simplehuman Sensor Mirror Pro
Amazon Echo next to Simplehuman Sensor Mirror Pro
Photo: Simplehuman

I’ve been coming to CES since before The Verge even existed, and in all that time one problem has vexed consumer electronics companies trying to build the smart home. That problem is interoperability. My smart TV, your smart lightbulbs, Samsung’s smart fridge, August’s smart lock, they all work in slightly different ways, use slightly different software, and don’t have a common language to unite them. And at this particular CES, I’m coming to wonder: why not just use human language? Let the gadgets speak to each other.

The device that inspired this idea in me is Kohler’s new voice-activated smart mirror, which can do fun stuff like dispensing a precise amount of water from the tap connected to it. The mirror already has Amazon’s Alexa built in and responds to your commands accordingly, but do they have to be your commands? Why can’t it be, for instance, your Google Home Mini conveying your instructions from a more distant place? We call devices like the Google Home and the Amazon Echo smart speakers, but so far we’ve mostly tended to use them as smart microphones.

I’m not entirely sure I’m serious about this idea, but I also don’t see much standing in the way of allowing smart home devices to communicate by voice. The proliferation of voice controls is so rampant that we already have voice-activated trash cans, which basically means that the capabilities required for this “new” communication method are already built into the latest gadgets for the home. The humorous way to look at it is along the lines of Toy Story, where your smart speaker and your fridge chat while you’re away:

“Sup, fridgey, how’s that casserole looking?”

“Not bad, Googley, still edible for another few hours. Is the Roomba still doing the rounds?”

“Yeah, it got stuck around the nightstand, but I helped it out.”

The more substantive matter here is that, once I get over the privacy concern of having Google know everything about my life, I’ll want it to also know how the contents of my fridge are doing, and to convey that information to me. There are smart Samsung and LG fridges that can provide that info, but there’s no easy way to plug that into my chosen Google ecosystem. Of course, I know everyone’s working on common APIs to allow home machines to communicate in machine (rather than human) ways. And it’s possible that everyone will just integrate Google Assistant in everything and make this a non-issue. I get that. But such efforts have been in development for decades now, and somehow there’s always someone that decides to Zagbee when everyone else is going with Zigbee. Voice is just a very easy, platform-agnostic way to hook up the tech in our homes that matters to us.

Much of my interest in this stems from my use of the Google Home Mini over the past few months. Somewhat like Twitter, I used to think smart speakers are too basic and limited in what they could do, but now I find myself relying on them with increasing regularity. Google’s voice recognition is especially impressive, and I think the overall progress the tech industry has made toward allowing machines to understand natural-language commands is often overlooked and underrated.

But smart speakers remain limited in their reach around the home, at least for now, and that’s where I think being able to talk to other voice-activated devices can be a benefit. For example, I’d love to be able to tell the Google Assistant to pre-heat the oven and remind me in 10 minutes when it’s ready. The fact it might do that via human language is endearing to me, and I think it could even be beneficial for less techie people, informing them about exactly what’s happening around their smart home.

Others might find the whole idea of audible communication between machines unsettling and/or weird, but I don’t propose forcing anyone into this. Nor do I think voice would be a very good system for controlling home security devices. And yet, it’s a version of the future that sounds a lot more fun than the present, seemingly interminable trudge toward common protocols and standards.