Has anyone been able to use the voice sound to locate speaker. I want to make my robot look at who is talking. I know this may seem complicated but it seems to me you should be able to us two or more microphones to locate and run ezb script for movement.
Ellis
USA
Asked
— Edited
3 Participants
Just use a couple of cheap analog sound sensors on either side of your robot... Read the analog ports and you will be able to quickly determine which sound sensor has the higher sound level.... Then move the "pan" servo of your robot in the direction of the sound sensor with the highest sound level... You can even add a 3rd sound sensor in the middle of the robot as well....
I would even add facial tracking to this so that when the head turns, it can then lock onto the face and follow it when the center mic has a higher reading than the side mics. Once a side mic has a higher reading, turn off facial tracking and turn the head to that side. Once the middle mic has the higher reading, turn on facial tracking again.
Great ideas. My robot has pan and tilt neck and eyeball and eyelid pan and tilt. I am going to design it to turn horizontal pan by sound and fine tune eyes an vertical pan by face tracking. I may need to get help on the interface script but I will try to do it first. Thanks for all the help.
@Ellis.
I just saw your thread. I've been playing around with sound sensors recently wanting to do exactly what you asked, and with recommendations and a script example that @mtiberia supplied, they are working great. So good in fact, that I have just installed a pair in to my K-9 build.
Here's the link to the thread that it was being discussed in that may help you.
Thank You for the info. Will check it out.