I have a simple setup where a master Rpi controls some components over wifi. What I want to do is to extend it to have voice recognition feature so that it can turn on and off using voice commands. Although I can use some hardcore voice recognization ways with a microphone like in this [link][1], but I want to use Amazon Echo to do it. But Amazon echo communicates only with WeMO which means I can't use it for general purpose. Now I understand that we can emulate WeMO like in this link, but I don't want to use it due to its limitations. Is there a way to somehow connect with amazon echo directly? Or manipulate it work for general purpose applications?
To communicate with smart devices, Amazon Echo advertises request on the network to find out relevant devices and if it found one, it sends them commands as explained in link 2.
There is another way to do it using MQTT but it too requires SQS http:// anrg.usc.edu/ee579/spring2016/Alexa/
[1]: http://www.instructables.com/id/Raspberry-Pi-Voice-Recoginition-Home-to Automation/
