Hacker News new | comments | show | ask | jobs | submit login
Google Home now supports multiple users (blog.google)
123 points by stedaniels on Apr 20, 2017 | hide | past | web | favorite | 46 comments

"[...] we ask you to say the phrases "Ok Google" and "Hey Google" two times each. Those phrases are then analyzed by a neural network, which can detect certain characteristics of a person's voice."

So two phrases is all this needs. Good to know. Allowing only registered users to use it might stop things like that burger king tv add that supposedly triggered Home.

Does this mean my voice can be used to finger print me? All i need to use is google voice function on my phone twice or maybe not even that if its listening? Seems like it. Thought I don't know what this will mean. Gait, sentimental and voice analysis seems kinda scary.

Android has had Trusted Voice for a long time where you train it against your voice by saying "OK Google" three times. The goal is to prevent unauthorized users from accessing your information while the device is locked. It's far from secure (Google says as much while you set it up). In my own experience my wife can trigger mine occasionally as well as just random noise from the radio or videos. I do get fewer false activations though so I guess it's better than nothing.

Maybe this new neural network backed detection for Home will improve upon Trusted Voice.

It's so secure it seems I can only talk to it in a small quiet environment at similar distance. In my car I get no reaction.

Maybe it's hardware related? My Pixel hears me in the car with music on.

I can't say. I've never used it extensively, but I'm tempted to say that it responded a bit better before I used the secure voice. I even rerecorded my voice, no effect. alas

It doesn't mean it can precisely fingerprint you. Only that it can fingerprint you enough to tell you apart from 5 other people.

It interesting, the person I'd say I sound most like in the world is my Dad. Humans often fail to tell us apart on the phone. I assume this is true for a lot of families!

I just hope any family member can turn off an alarm I have set. I could be out of the room or on the phone and they should be able to turn it off.

It will be decently trained on the wake word itself and probably not on anything else you say.

if you could change those words ("ok google", "hey google") to some passphrase it could help in many cases

Would be great to see this functionality in the rest of google's products, but I'm not optimistic. I'd love to seperate my kids' youtube and app usage from my own on my Android TV. Why is feature parity, or hell, even coordinated product direction, between platforms so difficult for google? Look at Google Home, Chromecast, Android and Android TV - They work together so poorly you'd think they were made by competing companies.

I say this not as an excuse, but as an explanation: these things are very, very hard to get right, and can be nearly impossible to retrofit onto existing systems. When every single one of your dozens (or at Google's scale probably hundreds) of interlocking backend systems designed at different periods by hundreds of different engineers were built on the assumption that one primary key == one account == one user, and you try to change those assumptions in anything more than a superficial way, it's a guaranteed multi-year project with a high chance of failure. Just look at the mess Microsoft is in with their accounts, also on the front page, as an example (https://news.ycombinator.com/item?id=14159085).

And of course, once you're a certain size (that Google passed more than a decade ago), you pretty much lose the opportunity to ever start over from scratch on this sort of thing. You're dealing with legacy systems forever.

Case in point: "to use this account with Drive, you must log out of all other accounts."

Can you expand on why one primary key == one account == one user doesn't hold true when you want to provide support for multiple users? In case of Google Home, each user has their own Google account and the device decides which account to use based on who's asking. How is the one primary key == one account == one user concept changing here?

You're right you can use completely different keys to access completely different account and it may not be that bad. But what happens when there's shared resources and none of your systems were designed for that?

The idea that comes to the top of my head would be controlling lights and appliances in the house. You don't have two completely different sets of lights and appliances in one house just because there are two users, it's the same set of things for both. But that means now you have to know that both users are sharing X while some other thing (such as a Gmail calendar) may be individual to each user.

Couldn't agree more. A (somewhat large) company I worked with started working on this tech about three years ago, and still haven't gotten around to even coming close to launching it.

Chromecast and Google Home actually work really well together. So does Chromecast and Android. Not sure what you're on here.

I don't have a google home, but I have android TV (NVidia Shield) and an Android phone, Android Watch and several chromecast audios. Google Assistant on my android phone and watch can't interface with the chromecasts or android TV at all. I can "cast" from any app on the phone to most of the chromecasts, but not via google assistant. Android TV can't send audio to chromecasts at all, and it's a special snowflake that can't particpate in multi-room audio sent from an android device.

I have something called "google assistant" on all 3 devices, and none of them have even close to the same capabilities or UX flows for media playback, so I basically don't even bother using it because it's just frustrating.

Do this everyday so not following?

Even my kids every morning fight over which Bob Burgers episode from Netflix they are going to watch with the Google Home (GH) and Chromecast.

Google home can do that. I think it's unique in that sense, where you cannot control other devices through Google assistant without it.

This is because not all Chromecasts are created equal. If you have a "normal Chromecast" it works great, if you have an Nvidia Shield that comes with "Chromecast built in" it does not integrate with Home at all.

Speaking of poor interaction, I have a Android TV (Sony Bravia 4k from 2015). If I want to connect to a wireless network, when selecting a SID on the TV, the only way I may enter the password to connect is with the Android TV app.

The issue is when connecting with the app, and attempting to enter a password, the app freezes, and then force quits. Making it impossible to actually enter a password to connect the TV to my wireless network. Luckily, the WPS option works, even though it's a bit hidden.

I've tried using the app with a Nexus 6, Nexus 6P and a Pixel! There is also no on screen keyboard for the Youtube app. The only way you can search is with voice, which is not the most ideal thing to do all the time.

As of the most recent update to the TV in January, casting has become rather intermittent. Generally I must hard restart the TV before my laptop or a my phone can properly connect to the TV. The Netflix app also randomly starts and takes over control of the TV even while in the middle of running another application.

I love my Android TV, but it's a bit absurd. I now have a Chromecast Ultra connect to my TV that I use most of the time, it also allows the Google Home to use the TV.

That's weird. I have a Sony Bravia 4K 2015 Android TV as well, and it sure has an on-screen keyboard, navigatable using the normal TV remote. Not sure what's going on inside your TV, but you might want to contact someone about that.

Apparently it's already been implemented in Nougat.


Internal divisions are pretty competitive, so you're not far off.

big company problems - each PA or even each team within same pa will have the overhead of a large company.

ps: google employee

This is great, but what I'm waiting for is for Google Home to allow me to have multiple accounts associated with a single user (Google Apps and Gmail).

I don't really see how adding multiple profiles for different users to the Chrome browser is related to adding multiple accounts for the same user on Google Home. Perhaps you could enlighten me?

Oh I'm sorry I completely misread your original comment!

No problem. Just wanted to make sure I wasn't missing something :)

If you look how Google has done multiple accounts versus Amazon you can see the core difference between the two devices.

The Echo has a code you use for different accounts and the Google Home (GH) just uses your voice.

The Echo is really more of a computer interface and has you do the work and the GH is intelligent and far more human in how it does things.

Really love the voice authentication with the GH as it makes so many use cases now possible.

For example, in our home I prefer some of my kids to be unable to lower the AC thermostat. Now I can have some able to when they ask the GH in the kitchen and others are not able to. But no awkward passcodes, etc.

The other is I am fine with guest able to do some things and then others I only want "privileged" users to be able to do. With the Echo it was trivial for people to learn the passcode. Now with the GH I say it and it will work and they say it and it will not.

But everything is like this with the GH versus the Echo. A huge one is the Echo has commands you memorize and the GH you just talk to it like a human.

So a little kid can use the GH as well as a grandma. Kind of like Google Search. Same text box for a 5 year old as a rocket scientist as well as grandma. Exactly how technology should be. Why on earth should we still have to use passcodes?

Amazon needs to replace the foundation of the Echo to have intelligence if they want to be competitive.

So long as the accounts aren't Google Apps (GSuite) accounts.

Wherein, only a crippled experience is offered.

I'm so ready to purchase Google Home and a Pixel device, but because I only have Google Apps accounts both are limited in what they can do.

It's great saying "it's the software and not the hardware", but when the software is crippled for a set of users, that's not great.

I'm slowly moving off Google's platforms after over a decade as a GApps user because of how nothing works for apps users anymore. Even if I could find a gmail address that wouldn't get buried in spam and that I actually liked, I'd have to re-buy all my electronic purchases because they can't be transferred. I might as well re-buy them all on iOS. I've decided that the tight integration google offers is actually an anti-feature if you find yourself outside the "one true use-case."

Google needs to seriously add additional wake words or phrases. Ok Google or Hey Google just sounds strange and just not natural. I have a GH and an echo. Compared to the GH, I find it easier to trigger the echo by simply saying "Alexa..." or "Computer, turn off lights". The "computer" wake word was recently added to the echo.

"Computer" is great even if just for the old-school scifi vibe it gives.

For best effect, pick up your desktop computer's mouse and talk into it like a microphone.

I know it's a tricky wake word but "Computer" is the winner for me. Now Alexa just needs to make authentic whizzing and beeping computational noises while it thinks

About time! This is a very important functionality for a device which is supposed to be in the living room. Glad they added the support.

I'm very surprised Amazon hasn't done this in the years the Echo has been out.

Great, and now let me add my kids and have some parental control features based on their voice recognition ;-)

I read that and the headline in my mind is now "cameras no longer needed to track people, can track people with mics only" ... "would you like some Gap Jeans" you were over heard talking about jeans at your local coffee shop this morning.

Maybe that's been possible longer. Maybe it doesn't mater with cameras everywhere. It's just a sign of where we're going.

PS: Not complaining either. It's a useful feature

I was really excited to try this out. Seems to be rather buggy though.

Every time I hit the "Multi-user is now available" button to set it up, a white box opens and then immediately closes. Even tried linking it to my SO's phone and configuring setting up multi-user from there, but it doesn't seem to be an option on her phone yet. :/

For the life of me I couldn't find the option again once I set it up on my phone. I just wanted to retrain the voice. However, when I did it from my wife's phone, the option came up immediately.

This is great. My toddler tried to make a Facetime call to my mother by saying "hey Siri, make a Facetime call to my dad's mom". Siri did not understand, since she assumes all voices are mine. I look forward to this feature becoming standard on all virtual assistants.

So does it support reminders yet?

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact