Last night was exciting. Roughly following the article at http://makezine.com/projects/use-raspberry-pi-for-voice-control/ I was able to get the sphinx speech recognition engine installed on my roadtest Raspberry Pi 3.
I'll save writing up this step to another night, but at this point I can say "Wow! I am impressed."
Grammar based speech recognition on the Pi 3 seems in a different league from earlier investigations with sphinx on my Pi B+. Now I need to bring my "old" Pi B+ up to the same state for real comparison.
Here is the first free text recognition:
a. b. c. d. yes g. h. i j. k. l. m. m. o. p. q. r. s. t. u. v. w. x. y. and z. one two three four five six seven eight nine ten
READY....
and this is the first grammar based:
Need more input:
I just heard you say:"TURN ON THE KITCHEN LIGHT"
I'm thinking now
Need more input:
I just heard you say:"TURN OFF THE KITCHEN LIGHT"
I'm thinking now
Need more input:
I just heard you say:"TURN ON THE BEDROOM LIGHT"
I'm thinking now
Need more input:
I just heard you say:"TURN OFF THE BEDROOM LIGHT"
I'm thinking now
Need more input:
I just heard you say:"TURN ON THE ROOMBA"
I'm thinking now
