Voice Recognition on Raspberry Pi

This is still a work in process. Each and every aspect of the process discussed below can and will be changed without notice until it is better refined...

This is a dual motivated project.  Voice recognition running on a Raspberry Pi without any need for cloud computing would be a great addition for a home automation system.  If it works as advertised, you can say "Honey, I'm home" when you come in the house, and your front entry hall lights can turn on.

Several OlyMEGA members have started a year-long project, where they're going to build HALO-style armor as costumes.  The HALO armor project is being discussed in the OlyMEGA forum  They plan on decking them out with lots of electronics (which I'll detail later).  I realized that voice recognition would make controlling the whole electronics package much easier.  Two birds with one silicon stone!

I've started with Jasper, and immediately ran into the typical tutorial strategies, where the authors give you 85 lines of new commands and configuration file entries without any explanation as to what the entries and commands do, or how the vaguely familiar selections were chosen for the specific application.  I admit that at this moment, I'm doing the same thing, but I am actively building the process (with lots of embedded comments).

Once stabilized, the code can be downloaded and executed simply (three lines in a terminal).
wget http://homeautomationhacking.com/c/jasper/jasper.s -r -O ~/jasper.s
chmod +x ~/jasper.s
sh ~/jasper.s

jasper.s will manage asking the user any needed questions to determine flavor/version/options of the configurations script.  It will then download the "actual" script which will perform the RasPi configuration and Jasper install and configuration.  Once the "actual" script is downloaded, you'll have the option to immediately run the script or delay execution so you can review what' the script will do.  On the server side, a PHP file will manage building and configuring the script that will download.