A.T.R. Audio Test Robot Above: The robot has a microphone on a short boom on the rear end, and the mic amp/signal conditioner is the small board on the other side rear. The eventual goal is for the robot to understand a small vocabulary of word commands. Updated 6/10/16 Key Search Words: ROBOT, ROBOTICS, ROBOTIC VISION, ARTIFICIAL INTELLIGENCE, AI

The next step in voice/sound recognition in our robot is to digitize the waveform for a period of time that will work for most words, and send that digitized result to the LCD display for viewing and comparison with the actual sound trace with the oscilloscope. 10 readings are taken after the trigger, at 50mS intervals to yield a total time of half a second recording time. This is JUST enough to resolve key portions of the chosen test words we will be using for our tests. The six words are chosen because they all appear very different in the sound traces and the robot should (I hope) be able to distinguish them! The six words are:

TESTING, FORWARD, ROBOT, CLICK, LIGHTON, LIGHTOFF.

The "Click" is made with your mouth as you might call a pet...it is not the actual word "click". This is actually the most rapid sound the human mouth can produce.

 The robot on the bench for audio testing and programming. The main processor is the large chip above center and is a PIC16F887 micro controller. On the lower right the satellite board is the audio amp and envelope generator. And to its right is the two line LCD display Ill be sending the digitized data to.

 Here is what a full second of the word "testing" looks like. you read 11 readings, from upper left to lower right in both lines. It would be nice to be able to send it right to a graphics display, but hey - those are not cheap!

 The scope camera is a web cam on a box that fits over the oscilloscope display.

 The word "TESTING" recorded for half a second on the scope. From left to right is half a second, the same time as the chip records the data. So you can directly compare the data with this image.

 Here Ive superimposed the sampled data plotted over the actual photo of the scope. You can see what the robot will see, its crude but just enough detail to differentiate words. It gets the basic shape pretty close.

 Some other words I recorded while I was at it. These are 1 second recordings of the KEY words. Here is the mouth "CLICK". The fastest click I can do is around 50 milliseconds....

 "FORWARD"

 "ROBOT"

 "LIGHT ON"

 "LIGHT OFF"
Previous Uploads on this robot:

   1.  The sound envelope waveforms with the mic amp board

    2.  Sound Activated movement sequence

3.  More waveforms, a scope camera and digitizing the shapes
  HOME