Built me a DIY talk box for electric guitar (ala Frampton, Joe Walsh, etc) but not terribly thrilled with it. Got to wondering about how to lose the tube that has to insert into mouth, and thought, is it possible to design a circuit that could reasonably mimic one? A Wah Wah pedal could be used to still affect the signal but in a way that the range of it could alter the sound more like a talk box sound. I know that it's pretty much impossible to get a sound and response to the depth of the human mouth, but I still wonder...anyone?

 

There are a couple of companies that make something but still lacking...

 

The "talk box" has to use the tube. It uses the volume of your mouth (not meaning sound level as volume) to make it work. Suppose it could be done mechanically using a cavity with a similar size and a motor driven diaphragm to change the volume of the cavity. Making it like a mechanical mouth. But don't think it would give near the same effect.

 

Complete fail. What you're saying is, "Look, I found the essence of what makes this work. I want to take that out and have the same results." That's like saying you want to take the variable resistor out of the wah-wah and get the same results. How about you take the blade out of your razor and the engine out of your car. Then you will get the same results...all 4 of them won't work.

 

Complete fail. What you're saying is, "Look, I found the essence of what makes this work. I want to take that out and have the same results." That's like saying you want to take the variable resistor out of the wah-wah and get the same results. How about you take the blade out of your razor and the engine out of your car. Then you will get the same results...all 4 of them won't work.

I understand what you're saying, but I still believe there is a way to mimic it, at least to some degree. Yes, I know how important the mouth part is, and I had been considering some way to build a small container-chamber and a moveable flap. I could see a linkage from the flap to a pedal that when used would vary the opening, simulating a mouth in a small way. (Yep, very odd looking I know but if it worked I could find a way to make it less unattractive haha) But then I started thinking the circuitry route.
I'm not trying to create the exact sound, but something similar to it. Since the Wah Wah pedal changes the sound in a way that approaches the talk box, why can't it take that same signal output and alter it more with the use of some variable components? What if I took an output of a Wah Wah pedal and routed it to another graphic EQ pedal that is set for a certain sound (think telephone speaker thin sound) and that EQ sound was brought into the sound, depending on the position of the Wah pedal? Even deeper, (my mind wonders yet again haha) what if a foot pedal was set atop a joystick that had at least four quadrants that when you rocked that pedal would send the signal towards different tone circuits (IC chips, etc.?) which would then change the original input tone?
I know I'm stretching, but that's what you guys do--you design/build new things. Electro Harmonix has a pedal with that in mind, and there are a couple of others (vocoders?) that try but I'm thinking someone out there has a better way, so I'm exploring. And challenging. I've seen a lot of great posts (answers) from some great electronics masters in forums like this. So I keep throwing out ideas...never know until you try...

 

Built me a DIY talk box for electric guitar (ala Frampton, Joe Walsh, etc) but not terribly thrilled with it. Got to wondering about how to lose the tube that has to insert into mouth, and thought, is it possible to design a circuit that could reasonably mimic one?

No. The spectrum shaping that happens between the vocal chords and the lips is too complex.

A Wah Wah pedal could be used to still affect the signal but in a way that the range of it could alter the sound more like a talk box sound. I know that it's pretty much impossible to get a sound and response to the depth of the human mouth, but I still wonder...anyone?

A 2-axis pedal can vary two parameters, such as the center frequency and bandwidth of one bandpass filter. But some of the filtering needed is phase-cancelling like a comb filter, and the multiple filters at work don't all change at the same time for the same reason for each vowel.

A DSP (digital signal processing) wiz probably could approximate what you want in a year of coding and tuning, but with one pedal it still will have only one or two variables where many more are needed.

ak