I want to work on my idea to do some modification to my BH3-R using volume of voice to control the speed of the walking robot. Such as saying ‘boo…’ to the mic it will make the robot walk a little faster when it is already moving, or saying ‘BOOOO!!!’ will make the robot go much faster.
Has anyone done this before ?
What I thought initially was just plain talking to the mic and have the other end of the mic plugged on to the BBII. Soon to know I was wrong. That is how dumb I am.
However I have talked to some people. I was told that I need to have a ‘voice recognition circuit’ to convert the voice to pulses of voltage ?
Has anybody done that before and has that circuit available ?
i have been many application were the user commands a robot to function vocally but i haven’t seen it responding differently or faster regarding the louder the command.
i will how ever say converting the voice to pulses of voltage is totally possible.
voice recognition circuit work as you may well know by converting the vocal variation to distinguish what word is being said and then “remembers” what the command is for the given pulse. so already having this i think it would be very easy to make it recognize the loudness.
That thread Kurt referred to is a circuit I am testing. I am at the stage where I’ve had prototype boards made and the first version soldered up. Working on a practical test now.
Here is a really simple design that I have used. You can play with the value of C1 to change the response time as well as adjust the pot for reference voltage.
I just googled to find more sound sensor circuit diagram and I found yours again on lynxmotion forum that you posted on 9 May 2009, viewtopic.php?f=26&t=4977&start=30
I suppose that this one here is the most updated one ?
But I notice that the MCU output from the previous diagram has only 1, while as the one that u have posted in this topic has 2 MCU outputs, may I ask where do I connect them on BBII board ?
Really great circuit you have here. I read again from other post you had it says no program needed ?
My project is to probably make my BH-3R to walk a little faster for 5 seconds while a ‘clap’ is sensed and slows down again. Is this doable ?
For the Atom Pro, I belive you use the output connected between R4 and Q1. It’s has been a while since I used this circuit but I think that is the one I used. Connect this output to any of the Bot Board adin pins. You will need to write basic code to create behaviors based on the output just like any sensor you would use. When I said no programing needed, I meant that the circuit operates without having to program code like you had to in my eirlier circuit designs.
My first circuit design used and ADC chip to compare voltage levels between each mic which required basic code to operate. This new circuit I designed, eliminates the ADC chip, and uses the comparator op amp to check the voltage levels. It works much better, is more sensitive, and less complicated than the older circuits. The newer circuit is very simple so it doesn’t do fancy things, but perhaps with some fancy creative code, one might be able to do things like find angles using time variations and other math related goodies. I have several schematics floating around but the one above is the most recent best performing design I have come up with so far.
Yes this circuit will work well using hand claps and you can write a basic program to count the number of claps that is given during a set time. The number can be stored as a variable and you can have different behaviors with each number. Loud environments may be a problem but a quiet environment will work well.
Did you mean LM741? if so, yes, an LM741 would work and yes 9v would run the circuit but it is better to use the regulated power from the bot board eliminating the need for an extra battery. Using an LM358 gives you the option of using two mics and it has low power consumption.
Yes the #1 output pin can be ignored. It’s used for Basic Stamp controllers.
I have been trying to build this circuit after trying for about a week. However, it has not been a success.
On a breadboard I build this ciruit, the components I have used differently are :
The electrolytic 1uF, I use a electrolytic 0.68uF instead.
The LM358, I use a UA741CN
The variable resister, I use a 22k resistor.
And from the output, which is said to be the node between Q1 and R4, I hook up a LED to it to test the functionaility of the circuit before I put on to the BBII.
I talk and blow to the electret mic, MCE-200 which I bought from Farnell, the LED does not light up.
If there’s a conversion of voice to voltage, the LED should light up.
I have already taken out everything and put it back once again but still not working.
Any insight on what might have gone wrong ?
I can post a close up picture of my the circuit if that might be helpful.
Make sure you look over everything. One wrong wire placement can result in failure. Pay attention to polarity, make sure you don’t have your +/- crossed. I made several mistakes bread boarding circuits, usually I always over look polarity or have something missing or connected or not connected. Check, recheck and triple check your work, and then recheck it again!
Hi, thank you for your reply. I shall re-check it at least 2 more times then. Meanwhile I just post the circuit here.
Just to test whether the circuit is functioning with an output between Q1 and R4, I can try with an LED right ? If there’s voltage then there the LED should lit up.
I have spent some time to tidy up my circuit and cut up unnecessary long wires. Hopefully that is clear enough to see.
The 2 black and red wires extended out from the board are connected to a 9V battery.
I am really desperate for some help here now as I have finished using all my own resources to solve this.
As suggested by SN96, I have taken out all the components, checked everything had it put back again. Checking its polarity and everything. I even had a senior lab assistant helping me out with the building of this circuit.
I have provided a regulated 9V power supply only then the LED from the original circuit lights up. A battery that has 5.5V left does not work.
Now the LED from the original circuit lights up.
The output to the BBII board is R4 and Q1 which is the MCU output 2 right.
Is this output here suppose to have different voltage when say, clapping and not clapping to the electret mic ?
When does the LED light up? Always, or just when speaking into the mic?
The output of this circuit will be as high as 9vdc. I hope you didn’t connect it directly to the Analog input of an Atom or Pro. If so the analog input will be damaged. You must limit the output of that circuit to 4.95vdc.
The LED from the circuit is always lights up when I supply a 9V to the circuit.
No I have not connected any outputs to the BBII yet.
If I have measured correctly, the output(which is between R4 and Q1) is always at about 0.72V, regardless of whether I have made any noise or clapping sound.
The nodes that I have measure is at the node between R4 and Q1 , and the node at the output of the operational amplifier.
The output voltage should vary with any noise made right ?
