The iPhone 4S

[moti barski]

some questions regarding siri's voice recognition :

1 is the algorithm based on words & sentences sound to text or phonemes to text ?
2 the sound data processed, is it the time between 0 volume levels(volume sample time) and volume peaks per volume sample time ?
3 what is the interval in which a sound (volume, buffer) taken ?
4 does the voice reco have a mistake reduction mechanism ? if yes, plz if possible, tell me some about it's algorithm
5 and most importantly, from where have the codes (or techniques) that are used were aquired ? and could they be used in vb.net ?

granted win 7 has voice recognition but it is difficult to train as the spelling recognition correction uses
the user voice, unabling the access to too many letters, at times