Audio Input for Voice Recognition
By Dr. Jim on Sep 3, 2009 | In Welcome
Dr. Jim has created an audio filter for his voice recognition software. It takes raw vocal data, amplifies it (first pic is the amplifier schematic), then passes it through a circuit he built that will take out the vocal cords, what he calls the "carrier wave" so all that is left is the modulation waveform. An analog to digital conversion takes place with this circuit after the carrier wave is stripped off.
He then is able to use this data, the modulation waveform, to pass to his machine intelligence technology so our robot can remember and learn from what it hears. We have effectively given our android synthetic "ears".

Here is Dr. Jim's video explaining this.
| « Dr. Jim Solicits Programming Talent | Voice Recognition Analysis » |