That was really funny

... althought part of the problem was that he was getting flusterred and changing the tone in which he was directing his commands. Ages ago I tried naturally speaking by dragonsoft or something like that and it was sensitive to pronounciation and I'm sure this is the same. To make matters worse, the quality of the microphone would dictate what the computer hears. Thinks like "S" and "F" have different high frequency content when said so if the microphone is not good enough, there is a chance that the letter S will not be heard correctly by the computer.
Anyway its a neat idea but definitly it will have to be set up correctly and the person dictating will need to be diciplined and not talk to the machine like its a human
