Profiling an acoustic guitar

    Hi!


    I wonder if this could be possible with the Kemper. I know it is not exactly what it was developed for, but I imagine it could still be possible.


    I would like to match/profile the sound of the pickup of my acoustic guitar with the sound recorded with the mic (in front of the acoustic guitar; no amplified&speaker involved).


    Is it possible to skip the test signal and just use the "refine" part of the profiling process to match the frequency and dynamic response of the pickup (straight into the KPA) and the sound picked up by the mic?



    Looking forward to hearing what you think!


    cheers,
    Karsten

    I should have formulated the title more precisely ;)


    Obviously, I did not mean to profile the acoustic guitar itself.


    Have you never wished that your amplified acoustic guitar would sound more like the sound that you record with a microphone?
    Isn't this exactly the idea of the KPA, to do some magic and provide us with the sound that is captured by the microphone?


    Think outside of the box. It does not really matter if the mic is capturing the sound from a speaker or from an acoustic instrument. The idea is very similar.

    Quote from klipping

    Think outside of the box. It does not really matter if the mic is capturing the sound from a speaker or from an acoustic instrument. The idea is very similar.

    I try to think out of the box and try to imagine how the Kemper is playing the guitar during the profiling process ;)


    The Kemper is sending Impulses and Sounds at one side and is analyzed the return on the other side.
    There are some vids which show the profiling process.

    Quote from Sharry

    I try to think out of the box and try to imagine how the Kemper is playing the guitar during the profiling process ;)
    The Kemper is sending Impulses and Sounds at one side and is analyzed the return on the other side.
    There are some vids which show the profiling process.

    That's why I wrote in my first post that it would be necessary to skip the test signal and use only the refine process which compares incoming signal, generated by a guitar pickup, with the mic signal.

    Quote from klipping

    That's why I wrote in my first post that it would be necessary to skip the test signal and use only the refine process which compares incoming signal, generated by a guitar pickup, with the mic signal.

    You should probably leave it to the Kemper people how they organize their profiling.
    Take it as it is. The Kemper is profiling amps in a certain way and nothing else.

    My first post - Yay!


    Klipping, I think this is a very valid question, and one which I have been thinking about.


    I'm sure you are very aware that there are a couple of acoustic simulators already profiled.


    Unfortunately my Kemper is in for repair at the moment and barely had a chance to use it but I do wonder what would happen if one were to 'loop' the out and in of the profiling process i.e. create an ultra-clean profile and then refine using mics and Di's, as you suggest above. I'd be interested to hear the results.

    Welcome @androo


    I’m really glad somebody else thinks that this would be extremely helpful and probably within the capabilities of the KPA.
    I will try that as soon as I have time.


    I have not tried (profiles of) acoustic simulators. Maybe I should give it a go.


    But I’m more curious to see if we can trick the KPA to take a snapshot of the acoustic sound and apply the KPA ‘magic’ to make the pick-up sound of your guitar very similar to what the mic is hearing.


    ...this concept sounds so familiar... :)

    Quote from androo

    My first post - Yay!

    Welcome, androo.


    Klipping, this sounds like a job for an EQ-matching plugin, mate.


    The Kemper relies on the Space Wars™, earth-shaking audio stream it passes through a device in order to Profile it. The algorithm/s is/are centred around that; it's its M.O., the result of Christoph's visionary idea.


    OTOH, what you're asking for requires the the unit to somehow take a snapshot, as you put it, of a device it cannot pass the necessary stream through. IOW, the fact that it's a Profiler doesn't in any way necessitate nor lead to an ability to capture devices through which it can't pass audio.


    We've seen this quite often. Folks think, "It's a Profiler, so... wouldn't it be great if it could do such and such?". Understandable, but when one looks at the method it uses, you can see that an entirely-different product is often what one is asking for, in this case, IMHO, an EQ-matching plugin is AFAIK the only available option out there short of a Variax or similar modelling guitar.

    I already amde a tutorial about this ... you can either buy a Tonedexter Pedal and profile the wavemaps, that works very well (i did that already)
    another workaround would be to use a match eq on your daw and route the kemper through your pc and then profile also the result of the match eq, that also works :)

    The refining process after profiling was always intended as a final correction and is no longer represented as important by the mothership.
    There might have been a change since the first versions.


    For strumming, appregios and also single note solis there exist useful profiles
    Can be found in the rig exchange or from professional vendors.


    I am afraid that we have to wait till one gear is invented to catch the sound of a classic guitar and it's complex fingering technique.
    For this a good micro technique seems to be the only solution at the moment.

    Quote from ReneLangenhan

    What about the fishman aura pedals?
    I mean you can still make IRs of an acoustic guitar and convert these with the cabmaker tool for the KPA.


    There are several ways of creating such IR files - just check YouTube.

    the problem with the aura system and my tone dexter profiles and the IRs online is that the source sound is different so they can sound good by accident but mostly i find them to be very off ...



    the tonedexter pedal and the match eq on the other hand would catpure the sound spectrum of your guitar and that fits perfectly afterwards ...

    Hi All. It's my first post here too, so I apologize if I cover well trodden ground.


    I can see two things that the kemper might do to accomplish what you are asking for, but both require jumping through hoops since the Kemper clearly was not designed for this.


    1. Linear response correction. This is done by capturing the impulse response of the system in question. This is usually done by exciting a linear system with a known signal, sine sweep, MLS, etc, but that can not be done easily on an acoustic guitar. That doesn't mean that you can't extract an IR though. If you have a pickup on the guitar, you could pluck a series of notes or chords recording the output of the pickup and the output of a microphone. You would then need to do a de-convolution of the dry (pickup) signal and the microphone signal. I believe the resulting IR could be loaded into the cabinet section, and would transform the pickup output into something much closer to the real thing. This technique would be much better than simple EQ as it would include time behavior, which is extremely important in an acoustic guitar due to resonant energy storage in the system (EQ is frequency space only, and does not include time behavior). This is basically a way to capture an accurate EQ with resonant behavior included, but still only deals with the linear response. In a real acoustic guitar non-linear response is what gives the guitar its actual tone, ie. what harmonics are included and in what proportion.


    2. Non-linear response correction. This requires actual profiling with the Kemper. I just bought my Kemper, and I have never profiled an amp with it, but I do understand some of the signal theory involved. I think it might be possible to use the built in profiling routines by exciting the guitar body with an audio exciter, such as this one. I have used this technique several times to study the resonant behavior of things like turntable plinths. The essence of what the Kemper is doing during profiling is making a map of the non-linear behavior of the system under test as a function of input frequency and amplitude. I don't see why a similar process could not be applied to the acoustic guitar. I can envision a few problems: One is that you assume that your excitation source, and its coupling to the guitar are linear. Compared to the guitar itself, this might be ok. The other problem is that this information will be duplicated somewhat on your pickup signal. Clearly your pickup already picks up some of the non-linear behavior of the guitar, or it would sound nothing like the guitar at all. If your "profile" of the acoustic guitar was taken with a linear source, but the actual signal it is applied to already includes some of this non-linear response I am not sure where that gets you. Perhaps if you can blend the amp signal with a dry signal it might still get more than the IR only technique. Or maybe it is just a can of worms.


    EDIT: I should note that the deconvolution mentioned in the linear response correction section might require some hoops to be jumped through in an of itself. I do not have direct experience with de-convolving arbitrary waveforms like this, but I think it should be possible.


    Sorry, I love making things do stuff it wasn't designed to do.


    Chad

    Edited once, last by CMHAudio: clarification ().

    Like 1
    Quote from Bommel

    the problem with the aura system and my tone dexter profiles and the IRs online is that the source sound is different so they can sound good by accident but mostly i find them to be very off ...


    the tonedexter pedal and the match eq on the other hand would catpure the sound spectrum of your guitar and that fits perfectly afterwards ...

    totally agreed. Fishman used to offer custom images of sent in instruments - but only for the U.S. market, probably due to shipping risks.


    I was more referring to the technology behind it compared to the profiler.

    Hi Chad,


    Thank you very much for sharing your knowledge. Very interesting.
    I have much less technical understanding and found your explanation was very helpful.


    Cheers!
    Karsten