"If you have more than one Echo or Echo Dot, you can set a different wake word for each".
This is something I've been thinking is becoming more problematic as well as an opportunity for real ubiquity. I have 3 separate devices nearby that are Google Now voice activated (the newer devices support this even if the screen is off), and they will sometimes trigger at the same time accidentally.
Since the processing is cloud based, and they know my identity, why don't the devices recognize this fact and cooperate. Instead of just 7 beam forming mics in the Echo, if you have two within hearing distance you could have the benefit of 14 and a unified response. Don't tie the request & response to a particular device, instead think of it as ubiquitous network that moves with you as you walk around the household, you should be able to continue your conversation from one room to the next seamlessly.
Since the processing is cloud based, and they know my identity, why don't the devices recognize this fact and cooperate. Instead of just 7 beam forming mics in the Echo, if you have two within hearing distance you could have the benefit of 14 and a unified response.
The echo and noise reduction software that I'm aware of can't really do that in a reasonable fashion.
With current solutions, you've got one DSP that's receiving all the audio streams simultaneously, and they need to be exactly synchronized in time. Then, using basically pattern-matching, it figures out what direction the user's voice is coming from, and combines some/all of the audio streams together to eliminate environmental noise and make the speech as clear as possible.
To do this with separate devices, you'd want extremely precise time synchronization. Which is possible, but I wouldn't want to implement it.
The extra processing and synchronization would take longer, and delay input to the speech recognition engine. I don't think it would enhance the user experience.
Just have the Echo that hears the person best be the one that responds. So simple, and easy to implement. I honestly don't understand why Amazon hasn't fixed this yet. It's so fucking obvious.
Ah yes, the rally cry of the person not doing the actual development work... In my experience, rarely is _anything_ "So simple, and easy to implement".
> I don't think it would enhance the user experience.
Baidu trains the voice recognizer by adding all kinds of noise to the training data. I think it might be easier to do that than use multiple microphones. The neural net learns to do the difficult process of separation of useful data from noise.
I learned to not have the wake word be "Amazon" when I was watching online training for AWS. The Echo went nuts until I finally paused everything and changed the wake word back to "Alexa".
They really need to make it so that all of the Amazon Echos on the same network use a proximity algorithm to determine which one responds. Simply: The Echo that hears you best should be the one to respond.
I want to have an Echo in every room, and I don't want to have to remember all their different names!
> I have 3 separate devices nearby that are Google Now voice activated (the newer devices support this even if the screen is off), and they will sometimes trigger at the same time accidentally.
> Since the processing is cloud based, and they know my identity,
Interesting, so everything said in that room gets processed and potentially sent to Google for indefinite storage? What a 1984-style luxury.
For the Echo, at least, it has to use your home network, so you could pretty easily run a packet capture to see if it's ever sending audio out when you don't want it to.
It's a cat-and-mouse game: What if it only sent the clandestine information when it picks up the "normal" word? The point is you don't control the device or its software.
exactly why I think none of this is worth it (echo, google now, siri, smart tvs etc) - especially given the current applications of the 3rd party doctrine - you are giving up the right to privacy for everything that is said in your home.
I agree with this entirely. I've been waiting patiently for a way to add microphone distance to my Echo and this is perfect for that... except it doesn't work that way.
I am very much hoping they fix it in the future and add a software layer to combine/route commands with one single wake word.
It's also a bit annoying that the Android Wear version of Now doesn't work the same as the regular Android version. For example, the full-sized one seems much more flexible with wording, and supports listening in several languages at once, while Wear is limited to one language.
This is something I've been thinking is becoming more problematic as well as an opportunity for real ubiquity. I have 3 separate devices nearby that are Google Now voice activated (the newer devices support this even if the screen is off), and they will sometimes trigger at the same time accidentally.
Since the processing is cloud based, and they know my identity, why don't the devices recognize this fact and cooperate. Instead of just 7 beam forming mics in the Echo, if you have two within hearing distance you could have the benefit of 14 and a unified response. Don't tie the request & response to a particular device, instead think of it as ubiquitous network that moves with you as you walk around the household, you should be able to continue your conversation from one room to the next seamlessly.