First of all, refer to my Jasper Article if you wish to get a little background on my voice controlled assistant adventures.
Having successfully hacked together a working implementation of Jasper on one of my Raspberry Pi's, I was still wanting for a better solution. Primary issues being accuracy of voice capture/translation and feature set. Of course, the lack of features was in part due to my laziness and unwillingness to code up several modules for my various desires. I did manage to get a Phillips Hue module up and working, but that was fairly short on features.
Enter the Amazon Echo, as I'm sure most have heard of already. Not only is the hardware integrated into a sleek, small-footprint package, but it is also crazy accurate. The array of microphones is able to pickup voice from every direction, even over currently playing music or media. This was something my Jasper setup was desperately lacking. All-in all, everything just feels more refined and well-thought out.
This is in no means saying that Jasper was a waste of time. Much was learned in that project which has piqued my interest in diving into developing features of the Echo. For that project, it is difficult to make the software work on a wide array of hardware. Whereas the Echo has a very specific set of hardware to make the software pair with and function best.
All-in-all, I'm quite excited to see what developers are able to do with the Echo and am looking forward to it's use-cases expanding even more.
Subscribe to Cognipository
Get the latest posts delivered right to your inbox