The voice recognition feature is so useless for "Kenny's" in the Japanese lesson. Some of the so easy phrases but couldnt be picked up by the software while it could be for tough, long sentences by Sayaka?