I recently published a blog post containing my predictions for Apple’s WWDC 2016. I touched on some of my wishes for all of their current hardware devices and operating systems that run on them. What I mention here isn’t necessarily a prediction, but rather a hope – or a plea – for something that I think is drastically needed.

Whereas one of my long shot predictions centered around a new hardware device that acts as a speaker or hub for the home, and is completely governed by Siri, this request – I think – can be much simpler than that. I don’t think it requires a new device, though I admittedly have no idea about how complex it could be to make it work.

What I want is for when I say “Hey, Siri!” that only the nearest or current device that I’m using responds!

Right now in my house, I have an iPhone 6S, an iPad Pro, my wife’s iPad Air 2, and her iPhone 5S. Two of those devices (the 6S and iPad Pro) have “always-on Hey Siri,” while the other two have it active only when the device is charging. The iPhone 6S only responds to my voice, though all of the other devices will respond to whatever voice speaks the “Hey Siri” command. It is not uncommon for the 5S and Air 2 (the older two devices) to require more frequent charges, and so will be plugged in. All of these devices will frequently be in the same room. So, a worst case example of what can happen right now is that, using my own voice, I will trigger all four devices to respond to my “Hey Siri!”

This is not helpful. And it gives the illusion of there being multiple Siri assistants, when I don’t think that should be how we view Siri.

(And I imagine it only getting more confusing if, as rumoured, Siri comes to the Mac next. If the function works the same way as on iOS devices, then we might assume that “Hey Siri” will also come over, and will work on the Mac while it is being powered.)

First of all, the devices don’t all reply with a perfectly synchronized voice. There is always a delay on some of the devices. So, even if all of them are saying the same thing, the reverberation or echo effect I get means that the response is next to useless to me. Secondly, and to complicate matters further, “Hey Siri” does not have an amazing track record for accuracy, so more often than not, at least one of the devices will hear me wrong. What happens is that I suddenly feel like I’m in the middle of a room hearing multiple conversations, unable to parse what any one in particular is saying.

I am not a programmer and don’t claim to have any idea of what goes into making these extraordinary devices work as well as I do. But in my mind, here is an idea: have all of the devices that get triggered initiate some kind of rapid communication between them that designates one device as the “responder.” How this responder device gets chosen could happen in a few ways. If I’m using one of the devices at the time, then clearly that device could take command and override any of the others. Alternately, if I just walk into a room and shout my command without actually using any device, then proximity could be assigned by how loudly my voice is detected. Nearer devices will detect my voice louder than devices on the other side of the room. All the devices can hear me, have a brief conversation amongst themselves about which unit heard me the loudest, and then only that device would respond to me.

I have no idea if this particular idea would work, but I know that there is a solution out there somewhere!

Leave a Reply