Got a vocoder & a fuzz?

[Paul Perry (Frostwave)]

I don't have a vocoder at the moment... but if I did, I'd be finding out whether it is worth running guitar to the fuzz, then thru the vocoder as carrier, and at the same time running the original guitar signal to the fuzz as voice. (not every vocoder lets you use two external signals).
Because, who knows, maybe that would result in an interesting "fuzz with reconstituted dynamics" effect.
And I know I'm not going to make an analog vocoder, that's for sure but maybe there is a cheap multifx vocoder to be had..

[Mark Hammer]

Not all that sure it would work great.  Thing to keep in mind is the two fairly different envelopes.  The clean guitar will die out fairly quickly while the fuzz will linger.  Certainly vocoding a fuzzed guitar sounds fine, but vocoding a clean one is likely to be frustrating.

Remember the old Gretsch Controfuzz.  That thing appears to have sound that gets fuzzier as it goes along because the clean signal starts out dominating, and by the time it has decayed, the heavily fuzzed side is still hanging around.  The end result is that it *seems* to go from clean to fuzzed over the lifespan of the note.  Very neat effect.  But it's neat partly because it has the very properties that would stand in the way of what you were thinking of.

I still want to eventually get around to my "talked to" pedal: a not-quite-a-vocoder.  The plan is to split the mic input up three ways, with three independent envelope followers.  Two are preceded by filters, and one is full bandwidth.  Each envelope follower drives a different optoisolator or other control element governing a different parameter.  So, for instance, the centre frequency of a swept bandpass filter in the audio path might respond to the amplitude of a highpass filter, while the distortion drive in a clippping stage might be set by the overall mic signal amplitude, and a modulation amount or clean/dirty balance set by the amplitude of a lower mid bandpass filter.  The idea is that one could modulate several parameters in real time by change the pitch or volume of one's voice.  Like I say, not a vocoder in the sense of mapping voice spectral content to carrier spectral content, but using voice to redefine the carrier.

In principle, one could have more than 3 envelope followers, but my plan is to aim for 3 simply because I think planning out how one was going to alter the tone over more than 3 parameters with one's voice might prove a little too confusing.  Presumably, the centre or corner frequencies of any pre-rectifier filters could be changed to suit the voice of the user.  I think it's a neat idea and if anybody felt like pursuing it and making it a commercial product, you have my blessings.  Zach, if you're listening, it's about time to set the 8-stage sequencers and foot probes aside and move onto voice control.

Merry Christmas one and all.

[petemoore]

  Thank You Mark> Everyone! Merry Christmas  !!!
 

[Paul Perry (Frostwave)]

Mark, I like the voice control idea! Plus I'm sure it would be a way around the patent that covers simply videoing the mouth shape & working off that.
Maybe if you were to make a number of "control sounds" and check the corresponding spectrograms it would give a good hint as to how to extract information from the voice. Pretty hard to get more than two independant control signals out, at first looking. Absolute level is obvious, and so is ratio of high band to low band.
Another problem...... unless you are into circular breathing, there are going to be gaps!

[DDD]

Talk-box sounds very close to vocoder. At the same time it's very simple and easy-to-control guitar gadget.

[snap]

talkbox is post-amp!

[GFR]

I don't have a vocoder at the moment... but if I did, I'd be finding out whether it is worth running guitar to the fuzz, then thru the vocoder as carrier, and at the same time running the original guitar signal to the fuzz as voice. (not every vocoder lets you use two external signals).
Because, who knows, maybe that would result in an interesting "fuzz with reconstituted dynamics" effect.
And I know I'm not going to make an analog vocoder, that's for sure but maybe there is a cheap multifx vocoder to be had..

My old Boss SE-50 has a vocoder. I have tried this idea long ago, I didn't find it good at all.

[Mark Hammer]

I don't have a vocoder at the moment... but if I did, I'd be finding out whether it is worth running guitar to the fuzz, then thru the vocoder as carrier, and at the same time running the original guitar signal to the fuzz as voice. (not every vocoder lets you use two external signals).
Because, who knows, maybe that would result in an interesting "fuzz with reconstituted dynamics" effect.
And I know I'm not going to make an analog vocoder, that's for sure but maybe there is a cheap multifx vocoder to be had..

I misunderstood you.   I thought you wanted to modulate a clean guitar with a distorted one, when in fact you had the reverse in mind.  Certainly THAT would work just fine.  Of course, some of the same caveats apply, and that is the relative duration of clean notes, versus distorted ones.  I think the other consideration is the relative spread or span of harmonic content in the fuzzed signal as a function of note pitch.  In the absence of complex filtering, bass notes have more upper harmonics represented than higher notes do (think of how many multiples of 1000hz fall within your hearing range as opposed to multiples of 200hz).   For reasons I can't quite articulate, my gut sense is that modulating a fuzzed guitar with a clean one might need the drive in the individual envelope-follower sections tracking the clean guitar to be adjusted a bit such that some passbands have an advantage.  The reasoning is that the harmonic content of the fuzzed signal would vary much more depending on how hard you pick than might the harmonic content of the clean signal.  Muddled thinking maybe, but again it related back to the notion that some of the temporal properties of the fuzzed signal change differently than they do in a clean guitar.  Heck, maybe the adjustment to be made is not in the gain of the envelope follower sections, but in the lag.