Unless you’ve been living under a case of 1 farad capacitors, you’ve heard of the Amazon Echo. Roughly the size of two cans of beans, the Echo packs quite a punch for such a small package. It’s powered by a Texas Instrument DM3725 processor riding on 256 megs of RAM and 4 gigs of SanDisk iNAND ultra flash memory. Qualcomm Atheros takes care of the WiFi and Bluetooth, and various TI chips take care of the audio codecs and amplifiers.
What’s unique about Echo is its amazing voice recognition. While the “brains” of the Echo exist somewhere on the Internets, the hardware for this circuitry is straight forward. Seven, yes seven microphones are positioned around the top of the device. They feed into four Texas Instrument 92dB SNR low-power stereo ADCs. The hardware and software make for a very capable voice recognition that works from anywhere in the room. For the output sound, two speakers are utilized – a woofer and a tweeter. They’re both powered via a TI 15 watts class D amplifier. Check out this full tear down for more details of the hardware.
Now that we have a good idea of the hardware, we have to accept the bad news that this is a closed source device. While we’ve seen other hacks where people poll the to-do list through the unofficial API, it still leaves a lot to be desired. For instance, the wake word, or the word which signals the Echo to start listening to commands, is either “Alexa” or “Amazon”. There is no other way to change this, even though it should be easily doable in the software. It should be obvious that people will want to call it “Computer” or “Jarvis”. But do not fret my hacker friends, for I have good news!
It appears that Amazon sees (or had seen all along) that home automation is the future of the Echo. They now officially support Philips Hue and Belkin WeMo gadgets. The Belkin WeMo, which is no stranger to the hacker’s workbench, has a good handle on home automation already, making the ability to control things in your house with the Echo tantalizingly close. See the video below where I test it out. Now, if you’re not excited yet, you haven’t heard of the WeMo Maker, a device which they claim will let you “Control nearly any low-voltage electronics device“. While the WeMo Maker is not supported as of yet, it surely will be in the near future.
We know it sucks that all of this is closed source. But it sure is cool! So here’s the question: Is the Echo the future of home automation? Sure, it has its obvious flaws, and one would think home automation is not exactly Amazon’s most direct business model (they just want you to buy stuff). However, it works very well as a home automation core. Possibility better than anything out there right now – both closed and open source.
Do you think Amazon would ever open the door to letting the Echo run open source modules which allow the community to add control of just about any wireless devices? Do you think that doing so would crown Amazon the king of home automation in the years to come?
“While the “brains” of the Echo exist somewhere on the Internets, the hardware for this circuitry is straight forward”.
OK, first of all I have to get my head around why any would let a global corporation put seven microphones in their home to stream everything they said…
Same thoughts here…
Correct, why aren’t there any cameras… I mean really !
And yes, that’s sarcasm.
Doesn’t need it – it uses 3D UWB-RADAR imaging. Probably there are versions for markets where people are interested in that the company song be performed with the proper reverence or the mandatory morning prayers be performed while assuming the required positions.
Seem crazy, let’s hope that this can spark development of improved free voice recognition models. I am for the technology but never by uploading audio recordings to closed servers.
Seems stupid. That may be why I’ve been living under a case of capacitors and never heard of it before.
You must of got zapped by those caps! The most remarkable thing I own! Much cooler than SIRI! It only records what you say after you say the wake word ‘Alexa’…I don’t care if big brother knows that I asked ‘Alexa’ to turn on my kitchen light! Or, if I added milk to my shopping list.
It’s not just about milk on shopping lists…
How does it know that you said “Alexa”? Presumably by listening to everything. Perhaps it only uses the cloud only to decipher other speech, and handles recognizing the one word “Alexa” locally. Or not. But would you really know if a firmware update (or config flag) caused it to (without visible indication) upload everything it heard for analysis? You may not care, but it concerns some people. To each their own level of care.
Cost benefit analysis. People in general don’t give a shit about their privacy. The complete lack of care as governments walk over people’s rights world wide, the wide spread adoption of smartphones which already do what this device does, and then even going back a long way to history where people would let companies install systems in their home to track their TV usage, or answer surveys in exchange for a food voucher shows this.
Okay, I can understand most of this: but “answer surveys in exchange for a food voucher”? How is that remotely the same? There, you’re choosing, explicitly, what information to give in exchange for an stated benefit.
That seems like the *right* way for companies to gather information, as opposed to surreptitious monitoring.
Agree. When you answer a survey for a food voucher, you are winning something and are free of what information you will share. For this kind of device, you are buying it to spy on you and the company behind it is winning twice (information and money).
I know its contrary to what everyone else is saying… but if you’ve got nothing to hide or worry about ‘them’ hearing.. I don’t see what the problem is. Sure it may allow them ‘access’ to parts of your life…. but well.. then just make sure they get an earful and bring it into the bedroom at night where you have sex or something.. let them get an ‘earful’ of that and see how long they spy on you.. Or just mess about with them and say all sorts of ‘keywords’ around it non-stop… like terrorist, cocaine, isis, isil, etc.. so they’ll get all sorts of ‘noise’ and make their ‘spying’ useless.
So jonathan herr, you are an insignificant person you say with not enough ego to even care about your own person – yet assume we care about your insane ramblings? Why not talk loudly in one of the many microphones you had you government install instead?
Jonathon Herr:
Study the speeches of Martin Niemöller: “First they came for the Socialists…”
http://www.ushmm.org/wlc/en/article.php?ModuleId=10007392
You may realize that having nothing to hide will only protect you for so long, and then it will be too late…
Whatnot, the only person rambling here is you, with your eager-to-insult-someone-with-a-differeing-opinion attitude. I dont even agree with him, but I still had to call you out on your senseless rambling
And yes… I stole your identity, because I was spying on the username of your comment
“People in general don’t give a shit about their privacy. ”
As long as the trains run on time, right?
Get over it, if the government wants to know what your doing or talking about, they will find out! Its not that difficult.
To not take advantage of technology because your paranoid, is only your loss. If you have a cell phone, or a phone at all, then you already are compromised! If your dealing drugs,or if your into other nefarious activities, then don;t talk about it, loose lips Sinks Ships… Otherwise embrace technology, its here for us to enjoy!!!
You are treating this as binary rather than a spectrum. If you are one of the top few targets of a major government, sure they have lots of ways to do surveillance on you. The key point is that they cannot afford to do that to tens of millions of people at once, nor keep it covert – unless said people pay for their own bugging and just trust that the microphones that no longer need to be hidden, will only be used for “good purposes”.
And it’s not only governments that want to know more about you.
You are welcome to trust what you trust, and use what technology you wish. Others will find their own level of tradeoff. That level of freedom of choice still exists.
Big Brother is watching you (well in this case he is listening…). Does this thing listen all the time or only on request?
I am sure its always on, otherwise it would miss most of the times you are actually talking to it….
At least, that was the case with the recent issue with the smart TV’s, correct?
It likely converts the speech to text, then the text is sent to a server at amazon to analyze, then it sends back error messages, text to speak, or instructions for the firmware to carry out. Speech recognition software could easily run on that device. Why would they constantly send streaming audio to the server? They wouldn’t be able to handle the bandwidth. Good speech decoding requires a lot of filtration, but it also requires a good sample. Compressed audio wouldn’t work.
You guys are all nuts if you think they are listening to everything you do. The only thing that it might do on their end is flag you if you say a lot of things that could jeopardize people’s lives–like making remarks about setting a meeting with Mark to go over the details of the upcoming school shooting you were planning to carry out. I’d implement it if I sold a device like this.
I believe it sends the actual recording to the Servers to do the Speech to text. That is the hardest part, if it did that locally, then all it would have to do is google it , then convert the response from text to speech locally. Much easier and better to send it all off to large fast servers .!!
It listens all the time, but I’m guessing it does not do the magic word recognition online. Therefore, only what is said after the activation word is sent to amazon.
Correct me if I’m wrong, I did not do research but have worked on development of utterance activated speech recognition systems.
Well, offline recognition engine or GTFO (c)
Today we put these in our homes, tomorrow lawmakers pass on a bill that will allow them to use those to ‘prevent terrorist attacks’.
Or for broad “legitimate law enforcement purposes” without adequate notification or supervision. When you build out a technology which is perfect for total surveillance, will it go unused forever, under all corporate and governmental transitions? Of course, it could just be for targeted advertising at first…
The voice recognition is offline while idle, and listens specifically for the wake word. Once woken, it streams phrases to Amazon’s servers in the same way that Siri and Google Now does.
But it’s closed source, and most likely has a remote update feature, so:
– you cant be fully sure of what does and what not
– if it doesn’t do it today, it may do it tomorrow
To me this and voice command for smartphones really look like a big brother technology. Should we start a “no voice recognition” ban like it happened with google glass?
+1. CMU Sphinx looks like the only choice we have.
+1. Add Julius as well
Wow! What are you people so paranoid of? I could give a crap who knows what I say or ask Alexa. If I didn’t want them to know something, I wouldn’t say it… Is it that you can’t help what you say?
A “No voice recognition” ban would only hurt us, the consumers. Big brother can put microphones into anything, TV’s telephones, light switches, AC outlets, etc. It has nothing to do with voice recognition. Embrace technology, don’t be afraid of it. Let it work for you, not against you! Of course some people will try to tell their problems to anyone or anything that will listen. If you are one of those, try telling a tree! They won’t reveal what you have said!
We have all had devices in our pockets for years that have multiple microphones in them and that are all ways connected to the internet… put your tinfoil hat on.
Speak for yourself. (Lovingly looks at Nokia N70)
Absolutely the coolest thing I have ever purchased!! With so many possibilities, Much cooler than my iphone, ipad, drone and other assorted gadgets! I have spent the last 2 days, since I received it, playing with it to see how smart it is. Ran out and order a few WEMO switches and a hue lighting system. They work awesome together! Also IFTTT
at IFTTT.com (if this then that) . Echo now has an exclusive channel with IFTTT. There you can configure many things. I can’t wait until more features are added. Like the API that should allow us to create our own features.
Way to go Amazon!!! I ordered this back in January 2015 and and got it May 20. Worth the wait..
I have one, and while the device is still admittedly relatively new (and thankfully supported by Amazon with ongoing updates), it is still currently a novelty. Right now, mine is basically relegated to a timer and an alarm. It’ll take some aggressive marketing from Amazon for it to become anything greater. They need to partner with more people, and take clues from other unrelated products. Those which may still be closed systems, but their manufacturers encourage hacking (two products off the top of my head are the Moog Werkstatt and the Korg MS20M)
well, it would be only a novelty if you only use it for a timer. It currently does many things that are useful ! Time of day, weather report, traffic, Jokes, News, turning on and off lights and appliances. ‘WikiPedia anything’ and play tons of music on demand.!! To name only a few! Your lack of imagination is the only drawback!!!
So if I tell my light to turn on, that request is send around the globe, processed somewhere and send back to actually turn it on?
yes, just like google does on your phone for voice recognition.
Of course. But its different if you explicit use this voice command or if something is listening to you all the time.
Isn’t google always listening now with its “ok google” command?
only if you activate it.
Only if you or the NSA activate it.
My phone only responds to “ok google” when it’s unlocked. That being said, I have no proof it’s not monitoring the microphone when it IS locked.
Except we have offline as an option for Google Now, well you can download the components for networks less functionality to keel worming. Personally I’m hoping to try and get a proxy running on the Echo for using a smart phone or something with now on it. That or if possible see about getting Now to partially run on the hardware. I don’t see why we’d need it to run it on a remote cloud when modern arm CPUs are so powerful.
Right. And imagine the world worked a bit backwards: if instead we had done it like that for years and years, yelling at our lights so they would switch on, and now suddenly there would be a brand new gadget: a light switch! A dedicated switch to turn your light on or off. No more messing around with settings in your phone, no more worries about the internet being down and your lights not working. Just press a button. How cool would that be? Everyone would rush to kickstarter to buy one of those. Some “inventions” just make life more complicated :)
But they are fancy!
I don’t know why there is this aggressive IoT advertising. Even for professionals… I don’t see too many use cases.
I attended the embedded world conference and heard a speech about wearables (held by a big semiconductor manufacturer).
Conclusion: if you have a display the battery life is shortened, if you have a big display, its even shorter.
I was… speechless.. (some guys payed a lot of money to hear the speeches)
Thank you for raising my faith in humanity back from the depths. I had begun to think we are all just children being led around by shiney things.
yeah right, like the keyboard your typing this with. What happened to the pencil?
yep, whats wrong with that? Who gives a crap if the NSA knows you turned your light on or off or asked Alexa what time it is? Too many paranoid people!
That is the huge problem, if they only would know single informations like that it wouldn’t be a problem. The more information they gather, the clearer the picture and the more private it gets.
“I have nothing to hide” is a extremely dangerous attitude.
“If you think you have nothing to hide, you simply have not thought about it long enough”.
“the hardware for this circuitry is straight forward” not really, their isn’t any open source microphone arrays circuits and software/firmware, could be a great HaD project.
The processor + memory isn’t that great too, RPi2 or OdroidC1 would perform well better for local ASR.
What bugs me the most, is the incapacity of a very fine grained tuning of the whole speech recognition logic, where I could select which local/cloud ASR commands can be used, and to setup a nested query/response selection system as I need it.
Closed sources and only cloud ASR, no thanks.
This is an OK devices but its not open source and that’s a bummer. If you look the Homey from the company Athom. This device has a beautiful design and its open source. only its still in development, but will e released somewhere in the summer of 2015 in the Netherlands.
Holy cow the “WeMo Maker” is expensive. Isn’t there something cheaper that can be made WeMo compatible?
OK- so it being closed source is bad – anyone fancy stringing together an open source version? Lots of the components already exist or can leverage existing (closed) API’s (IFTTT for the final hop wouldn’t offend me) – it doesn’t seem like a massive stretch to think that I could use an old Android handset, bluetooth headset (Star Trek communicator pin badge anyone) to build something that could achieve almost everything the Echo does- or am I missing something?
There is a suit of programs intended to do just that, http://sirius.clarity-lab.org/. It is a combination of speech recognition (Sphinx), question-answer (OpenEphyra) and others.
Um, when Samsung’s smart TV was discovered to send voice data containing your private conversations to the mothership for processing, there was an enormous outcry (http://techcrunch.com/2015/02/08/telescreen/). And this thing is supposed to be “cool”? It is always on and always listening.
Sorry guys, but this are really only gimicks with the sole purpose to collect and analyse your behaviour data so that you could be sold more crap. Marketer’s wet dream. They also make a wonderful tool for burglars that want to see when you aren’t at home if the data is accesible online or able to be intercepted. How many security holes are in the proprietary software? In addition, these things tend to be unfortunately universally crap at everything else.
No, just a few odd people were complain about Samsung TVs initially. The internet wide outcry was that they were sending your data back and forth completely unencrypted and then were caught lying about it.
The online voice recognition is offline while idle, and listens offline specifically for the wake word. Once woken, it streams phrases to Amazon’s servers in the same way that Siri and Google Now does.
Suuuure it is. I have some shares in a bridge, Interested?
This kind of technology is taking away real human privacy.
Everything we do online is being tracked anyway.
Now mobile phones and websites are doing the same.
So were are looking at a sort of minority report future at this rate.
God’s Eye
-Fast & Furious 7
Exactly.
Who or what monitors these activities.
I mean someone has clearly stirred the soup for people to push recognition tech so hard.
As a final “product” I think that Echo sucks.
BTW hardware side is really cool and it will open a lot of different opportunities, it’s not a killing product due to these factors:
– closed source. It could be fine because they want to sell you something (a product, a toy, a sw service/paid subscription, …) but if I cannot integrate it or mix its contents with something else, now it is just a funny toy
– no APIs, internal docs or any additional features to develop something different with it
– no market/appstore/XYZstore, … If I wish to load something different on it there are no applications, developers, indie devs or someone else outside Amazon team with it
– no community. A teardown is always cool and interesting but if they realize they can do a lot more with it this could be a killer product. What about creating a community or disclose details with an open source group ? I think they’ll benefit from the community, a product must be part of something else if you don’t want to let him be alone in a corner. A Community may return something back to them (bugfixing, applications, new features). Do you want to sell your product ? let me know how it works at least… or the next big thing could be something different, something like a RPI2+*duino+fpga+HD audio with steroids.
In just one phrase: open source it !
Disclose internals or at least let it be accessible with your APIs and let us flash it with something different. It’s a nice device but I prefer to have my poor ASR, I already know that NSA ASR is far superior but I prefer to have a choice at least.
Amazon can you hear me ?
Good day Ben,
Apparently, there is an API, but you have to request it… although I am still waiting …
https://developer.amazon.com/public/solutions/devices/echo
Cheers
Hi Kirby !
I have seen that as well ! but I am still waiting for a reply like you :)
I was thinking about some sort of personal mistake !
“Echo, add to your TODO list: tell amazon to reply to users and disclose APIs”
Amazon didn’t respond to my request for access to the Kindle SDK back in 2011. It might be a pattern.
Yes there is an API, and a small community of people in the program. Best I can tell, Amazon is feeling its way here, working out what works and what doesn’t. Integrating lots of random apps is a tricky problem – you need to get the right app to respond to random things the user says, but without accidentally unlocking the house door or some other undesirable side effect. I think it’s a hard problem to solve well.
1) Why would I buy a device that listens to all that is said and sends it to who knows where?
2) It is MONO sound. Yes, like in the 1950’s. And does not have stereo output (i.e., to connect to a better audio system) and does not pair to external BT speakers either. Extremely limited if you ask me.
3) All for what? So I can ask amazon about the weather?? Really?? Or to listen to MONO music? (as in 1950) Really??
4) And you want me to PAY for that???
If you consume pop music almost all of it is mixed down to mono, so there is not a great loss, not that a tiny speaker in a cheesy plastic enclosure will give much sound quality, be it mono, stereo or crammed with speakers.
WTF? I think your stereo system is busted if you think all pop music is mixed down to mono. You can say a lot of negative things about it, but that’s not one of the things.
So many people have posted this. So many people haven’t even bothered to check if this is even true. The Echo listens with local voice recognition for the wake word “Alexa”, and only while woken and receiving a command does it send voice recordings anywhere. Just like Google Now and Siri.
I’m a fan of online voice recognition (after a wake word), because online voice recognition always has way better accuracy. Not to mention they can add new features on-the-fly instead of “Upgrade to firmware 2.2.3 to get X new feature!”
we say it because we understand what you’re saying is a complete BS answer and could easy be subverted by both hackers companies and governments to listen for any set of words with a few key strokes…
the point is you only have blind faith that some ones hasn’t changed the code and isn’t spying on you and this is why online voice rec is trash.
Greetings fellow citizen, we would like you to pay for and install this bugging equipment in your home, and to save us some money, can you please power it for us – thanks you very much. Your help is greatly appreciated by all us folks working in the 1984 department of the NSA. It is only through everyone’s vigilance that we prevent the terrorists from taking away out freedom.
I know that the Xbone one guarding the freedom of your living room with the 3D infra-red cameras was loved by all, but the audio quality was just not good enough. This new device with it’s phased array of seven microphones can help us track people as they move through your home. With enough data we can hone our keyword detection and dispatch our collateral protection drones to well err, protect you, yes protect you.
We installed an alarm system several years ago. It has a microphone and speaker, when the microphone detects one of the remote (not wired, not wireless) alarms going off, it calls the control center and they respond through the speaker “We have received an alarm, is everything alright?” (or something similar). I was surprised when the installer walked around the house, speaking to the unit, and it responded even when he was in the basement near the other end of the house. This feature is to detect the voice of someone trapped in their room by fire, or has fallen and can’t get up [TM] to alert emergency personnel that someone is inside needing rescue. If you get this Echo, don’t assume that it will only pick up conversations in the room you set it in.
Back when I had a 1.2GB Quantum Bigfoot HDD packed with voice recognition software like IBM’s ViaVoice, I could voice control my computer (quite nicely, actually) WITHOUT having to install what is essentially a 1984-esque telescreen. What did my old P6 powerhouse do that this can’t? IMO, the only way I would use a product like this was if it either processed locally or MAYBE processed it on my own automation server. Fortunately, it doesn’t matter if it is opensource or not; I’m fairly confident that someone out there will RE the crap out of this thing.
As a note, that system probably consumed over 10 times the power to do similar things. I’d *love* for more systems to have local voicemail recognition, but I’m not sure any of of them have the processing capability.
Voice recognition. See, they can’t even autocorrect well.
It may have been a power hog, but that was not a requirement of the voice recognition technology itself. It was merely the acceptable level of power consumption of the time. A Raspberry Pi nowadays can do the same exact thing much quicker with much less power. Embedded systems have more than enough power to be able to process these simple, discrete vocabularies with minimal training. Being able to quickly process arbitrary speech from anybody with little/no training as accurately as possible? That’s a completely different ballpark. It’s still irrelevant in this context, though. You don’t need a massive server farm to process “Living room lights, on” Hell, I vaguely remember writing a tiny Python script that used the Speech Recognition Engine that came with Office… That was a decade ago! Really, the only problem I really had was being able to capture audio clearly enough from anywhere in the room to yield accurate enough results.
I’m currently reading 1984. That telescreen comparison is very true.
Today I tried that Google Voice Search thing. And said something I hoped would mislead it:
“Ok Google, Led Zeppelin lyrics.”
Which it pulled right up. It didn’t think I meant LEAD zeppelin, like Dragon or Microsoft speech recognition would have. Or require me to correct or train the engine – usually multiple times – to get it right in the future. And if next year I find myself listening to a new rock band called “Neoprune”, and ask for lyrics, it will recognize it with no action required on my part – not mistake it for “neoprene” until I personally add that word to the dictionary.
That is what cloud-based speech recognition can do, that offline solutions can’t.
Why does everyone think it’s always on? You have to “activate” it like you would Siri or OK Google.
As i understand it, the activation is done by voice, sending everything it hears off to be analyzed, looking for the activation request. Siri / Google are manually activated, then listen.
The trigger word is handled locally. If you disconnect you’re WiFi and talk to her, she’ll tell you there’s no network whenever she hears her name.
Thank you! All these people posting don’t know what the hell they’re talking about.
and no one can hack it to listen to other words nope nothing to see here folks
If they could have it listen to another word don’t you think someone who is not a black hat would be posting it to hackaday to show off what they did? There are hackers out there trying to do cool stuff rather than just steal things.
black hats aren’t the only people out there that hack things Pablo, the NSA has already shown how much they love putting back doors into devices like this. honestly this is just asking to get spied on for little to no advantage really.
dex- the NSA has already been listening to phones and stealing email. If we change our behavior based on what the NSA might be listening to we might as well stop communicating entirely.
How does the trigger word work ? Is it a special chip that wait for that word and then send an interrupt to wake the CPU ? Or is it the CPU always waiting for it, sending data only when the trigger word is pronounced ? In both cases, how do we now that there isn’t other trigger words that would send data but not respond back to the user ? Or simply sending data without waiting for the trigger word ?
Because it’s always listening to you (if it wasn’t, you couldn’t trigger it with a spoken keyword) and the voice processing is done off board. Streaming data constantly is necessary for the system to work as designed.
How would you feel if I told you that Facebook analyzes everything you type into a text field on their website, even if you later delete it and never post it? They have to, because that’s how they let you auto tag people’s names in your posts as you’re writing them, but it sounds a lot creepier out of context.
Alexa – Blow me !!!
Up pop the porn ads.
I have an echo. I bought it with the intent of it becoming a hands-off calculator/conversion tool. Other than that, as said above, alarm, timer, and weather report.
It would be a lot handier if it were in every room.
I’m pretty sure that Amazon has thought a lot about the objections above and what a PR nightmare it would be if it was ever proven to be listening at all times, and is very careful to make sure that nothing leaves the house without the keyword precursor. Not that it isn’t ripe for hacking but I would guess that’s why you can’t change the keyword – so that it can’t be easily hacked to a keyword you didn’t expect, like “bomb”
On the other hand, most of the stuff you say around your house really isn’t that interesting and your every online move is parsed already. If you really plan to overthrow the government then it’s likely you’ll be smarter than the guy who bought a lottery ticket, then hacked the lottery computer to produce those numbers, thinking nobody would ever figure it out.
Unplug it.
Correct me if I’m wrong, but I thought all these devices only send data to the cloud when they’ve been been triggered – i.e. listening and detecting the keyword “ok google” or Alexa or Siri is done on-board. Once the device is triggered though, it will send command audio to the cloud for processing.
Yes, they probably all work very similarly in this regard. Have anybody run wireshark/tcpdump to see what sort of traffic an Echo sends when it’s idle?
Monitoring transferred data can be a dangerous way to validate a device’s ability to eavesdrop. It can very easily lull you into a false sense of security. Even if it wasn’t encrypted, even if you could directly monitor the audio stream data, you simply cannot verify that it sends ONLY what it should. You can embed extraneous information into the data stream. Imagine if it only sent data back after it was ‘activated’, how can you guarantee it isn’t sending the audio when it was idle? You can’t. If something as seemingly benign as a laser printer actively hides information able to track you, what is restricting this device from doing the same? It requires absolutely no effort on our part to jump to the conclusion that this may do more than just what Amazon says it does.
worse, stateful logic could bite you, later. maybe things it hears and learns will combine with some other word you say later on and then, well, maybe not so nice things result from that.
anything that has a mic or a camera is creepy to me. unless there is a power switch that is simple electronics 101 style (not a software bit that is readable but a true power switch that works) I won’t trust it. I won’t have it in my house, either.
amazon, google, apple keep testing our limits. its surprising that we have given up as much as we have, already! ;(
BL, So you don’t have a cell phone, laptop or tablet? All of those devices have cameras and microphones. They rarely have a “true power switch” Even if they did, how could you tell they didn’t have a software controllable bypass on that switch without cracking the case open and following the traces on a multi-layer circuit board?
Yes, the coded information sent to submarines is filled with new stories, sports scores, whatever, to keep listeners from determining that something is going down (no pun intended) by an increase in traffic.
First let me say that I love home automation. I’ve done tons of things just to see if it can be done (not because it was a good idea) and I really want to see HA, Smart home and IoT take off. That being said there are some really strange things out there (this is not really one of them, but …)
Do you actually think your TV is off when you hit the off button ? You might like to grab a kill-a-watt and see (it’s not). Despite sounding sarcastic,I wasn’t trying to be this time. How can we place any trust in anything where we’ve allowed the outside (a Company that can control the device) in? One software glitch and suddenly more than you expected is being shared with the world.
My modern view of a publicly controlled company is that they’ll do what ever is in the best interests of their stock holders and that I am no longer truly a customer of theirs but rather a product, sold to the stock holders. Because of this, I’m very leery about the use of cloud services where I can’t secure the data.
“Because of this, I’m very leery about the use of cloud services where I can’t secure the data.”
This^
while I haven’t gone completely off the grid (quite a ways from it) I don’t store stuff on Dropbox or any other “cloud”,
service… (except my Norton backups, and I’m not sure those could be trusted).
I find it odd that people really want to talk to a smart home or home automation. Perhaps to get information. I actually prefer to have this stuff automated and need no, or little, user intervention as possible. The displays I have for things (which still needs work) gives me the necessary info with color, images and some text. I really don’t want my house to listen to my wife and I argue, it hear my wife to perform some impossible task and the TV suddenly turn to the adult channel to show me products and service that might make that possible.
Its such a bad idea to put our data in one place. You know Amazon will keep this data, the NSA and hackers will listen to it and abuse it. This is one step away from the thought police.
it’s packed with 7 mic’s so it can do echo-location , hence the name. doesnt need camera’s because it can “see” perfectly clear and not only in the room it’s in. that’s also why it’s packed with speakers … low freq beacon echo’ing your whole house. probably also monitors everything that happens on your Lan network just to be sure they don’t miss anything. So no i’ll stick with some opensource thing i know has faults people can fix and stays within the walls of my home.
What if you have one of these answering machines at home that actually plays what you are talking into the phone and you say things like:”Alexa, buy a new car.”. Would you consider this hacking? I would really love to se a youtube review where somebody tries this if it would actually work.
I didn’t know Hackday did advertorials now, how much do you charge to peddle closed source, hacker unfriendly junk from giant corporations?
Doesn’t the FTC regulate this?
Closed Source? What’s this? I didn’t dig into it but there’s > 1 GB of something there
http://www.amazon.com/gp/help/customer/display.html/ref=hp_left_v4_sib?ie=UTF8&nodeId=201626480
Set up three Echos under three different accounts:
#1 listens to FOX news 24/7
#2 listens to MSNBC 24/7
#3 listens to Soaps every day as long as possible
Interesting to see if adverts fed to the corresponding accounts feature guns, butter, and dildos. Straightforward way to prove out the targeted data thinking.
A few “Hey Bezos, making money yet?” will assuredly get your account cancelled.
For those of you that are interested in this device but are concerned about it listening “all the time” you do have a more secure option. The device has a button on top that will turn off the built in microphones. You can then grab the remote that comes with the device and use the push-to-talk microphone that is built into the remote to talk to the echo unit when you want. This allows you to not have it listening “all the time” and still get the benefits of the voice recognition.
It is really embarrassing that most of this discussion is centered around complaining – I thought this was a place for hacking!? Given that there IS an API available; whom has done some having with it!?
What are your opinion about Jibo ? It look promising and more interesting than Echo for me, personnally I can’t wait to getting mine and getting access for the developper api.
The echos of ‘it just like siri or google” are so obviously paid astroturfers that it hurts… LOL. This is your JOB? And I thought my life was pointless..
I’ve had my Echo since December, and it’s a great device. It definitely has potential for being a terrific HA hub/interaction device. While the fact that it’s closed source is somewhat disappointing, the fact that they are actively developing an API for it and expanding it’s functionality is awesome. I certainly hope they open it up and allow third party apps because I will definitely be integrating it into my custom HA setup then.
Actually I was mistaken. Amazon has released a great deal, if not all, of the source for Echo – http://www.amazon.com/gp/help/customer/display.html/ref=hp_left_v4_sib?ie=UTF8&nodeId=201626480
Reads to me like there should be ready market for an outboard firewall to prevent some remotely turning on the micro phones. Then again the NSA will finagle a back door in such a firewall ;) [sigh]
Hello. Does anyone know the part # of the power & speaker driver circuit board and where it can be ordered? Thanks.