#!Python

####################################
#for purely text-based applications#
####################################

# add the following lines to your imports:

from SpeakPython.SpeakPython import SpeakPython
from SpeakPython.Result import Result

sp = SpeakPython("[appName].db");
r = sp.matchResult(input);

if r == None:
    #couldn't match anything
else:
    rStr = r.getResult();

    #now do something with the resulting output string such as:

    #direct execution
    exec rStr;

    #simple output
    print rStr;

    #system control commands
    #(see SpeakPython/examples/LinuxCommands/)

    #parsing output for execution/post-processing
    #(depends on parse)

#!Python

#####################################
#for speech recognition applications#
#####################################
#taken from the HouseCommands example

from SpeakPython.SpeakPythonRecognizer import SpeakPythonRecognizer

#define callback function with 1 parameter (string)
def execute(s):
        print s;
        #exec s;

#creates recognition instance
#param 1 - function accepting 1 string argument - used as callback
#param 2 - appName - used to read [appName].fsg and [appName].db
recog = SpeakPythonRecognizer(execute, "houseCommands");

# sets the level of debug output
# 1 is the most output, 10 is the least
# default level is 3
#recog.setDebug(1);

#call this to start recognizing speech
#I believe this call utilizes current thread. Please multi-thread it yourself if necessary.
recog.recognize();