Rather then a single base unit placed in one room with an option of a remote that a user has to carry. Could you give us the ability to have a base unit that's hidden away with the option of adding sub-units that are basically microphone/speakers. Then have the ability to place wireless sub-units throughout every room of the house that's aware of the room its in. Weather it be labeled by the owner and add select controllable devices for that room. And able to handle multiple request from guest/family members of the home for their specific room.
To add onto the request, below is what I posted in another thread regarding multi-echo coordination. Think of Echo as Jarvis (stay with me here). Jarvis exists on one of Stark's servers, and Stark interacts with it on all of his devices. He talks into a microphone in his suit, which then relays the sound to Jarvis' server. Jarvis then computes the necessary response, and sends it back to Stark outputting through the speaker of the device Stark initially spoke into. Echo could work the same way. You could extend Echo's range by connecting a bluetooth speaker/mic to it (going forward I'll call this the Add-On). The Add-On doesn't need access to the internet, as Echo is going to do all of the heavy lifting. The Add-On microphone behaves the same way as Echo's - it doesn't start "listening" until you say the wake word. So to the end-user yes, the Add-On would behave "essentially" the same way as the Echo does, behind the scenes the Add-On isn't anything but a talking box. Echo would be the device that is actually processing everything. It would require that the Add-Ons be named (something that could easily be done/added to the Echo app). The command could be something like "Alexa, tell Kitchen it's dinner time." Then the Kichen Add-On would output "It's dinner time" through it's speakers. This is totally possible with the Echo. I unfortunately, have no idea how to implement it - but that's why we have Amazon reps in the forum.