Here in the States the Amazon Echo using the Alexis voice service has become very popular. My wife has asked if it would be possible to incorporate the Alexa functionality into my current RasPI based Bot. Amazon is offering access to its Alexa service and Voice services for free for non commercial use. 

I am looking at spending a significant amount of time to work out a soloution to get access to this capbility through an MRL service. Before I start I wanted to see if there was anyone else already working on this, as I would hate to duplicate effort. Or if anyone has any words of wisdom or guidance on this task. 

As you may know I am not a programmer and this is a massive stretch for my abilities, but the developer support material from Amazon seems reasonable and if it was easy it wouldnt be fun! 

Wish me luck! 


8 years 3 months ago

Great Idea BrotherBrown !

We would welcome it.  We love to Borg !

My words on how to proceed would be :

Step 1 - Research the Alexis services, update this post or make a new post describing the capabilites, good points, bad points, and list of references, APIs, documentation, previous examples, etc.. etc.  The post will be like a little island of info for all of us to share.

We will certainly help on subsequent steps..

I know very little about the Echo service .. these are the little tidbits I do know..

Amazon was about the first project I know of which potentially has the ability to get a "Borg" into many peoples homes...

Cynthia Breazeal a MIT deciple of Rodney Brooks (my hero) did some amazing work with social robots - she has her own similar product

They are all very "Borg" like - in that its all networked together with cloud services, background upgrades, data flowing both ways ..  creepy but cool...  changing our society in a way (similar to smartphones)

I'd be excited to see what Amazon can offer us..  But we will be in the same position as Google Speech in that although it's "free" its not Open Source (big difference) .. Most of the big players are all interested in moving to service providing vs product providing ..   Like the "service" of echo vs just a product..   

Its a bit of a double edge sword - in that its usually very good quality, but at the same time - they give you a watered down interface, or they can turn it off whenever they want.

Regardless, we are all interested in this concept, potentially it could be very powerful to have a "Cloud Brain" for InMoov or other robots to share, and perhaps it will springboard our own Open Source Cloud version.

Go forth and explore !  ... and tell us all about it !

GroG, you are correct in your assumption this is a free service, but not an open source offering .

Amazon is offering voice services for free, there seems to be alot of restrictions. Its seems simple enough to use as it uses a REST API, but only returns audio values. I will  conitue to look down this road, but it seems that something a little more open might be a better fit, perhaps the upcomming "MyCROFT" open source Echo challenger would be a better fit. 

Here are a few links to the basics of AVS and Alexa.

Getting Started Guide:…

API Service Guide:…

Reference Implementation guide:…


If anyone is interested in the MyCroft product, the SW stack is due for pre-release in Apr 16:…


Strangely (or perhaps not really so strange) - the first I saw MyCroft mentioned was after I became part of MaryTTS forum.  

It was a request from MyCroft developer on how to get a better sounding MaryTTS voice - (I don't do the question justice - but it was well formed and in-depth)

Anyway, just thought it was interesting as we (lead by MaVo) are continuing to try to make MaryTTS better.

Here's a Mycroft github repo ! 

HA ! .. ITS WRITTEN IN CLOJURE - WHICH IS JAVA "LISP" :D  .. a good match for us !  Sweet !