know any plug in that's capable to process sounds in real time (as I play) turning the sound into something completely different?

the basic idea is: say I have a flute and while I record I keep the motion, nuances, vibratos, in few words the whole performance
but the sound is completely different .. or at least it doesn't sound like a flute... same playing a guitar with bendings, hammer on, pull offs etc...

no choruses, phasers, flangers etc... are admitted :)

I'm looking for something different than the "usual suspects"

3 Answers 3


You can use an Audio to MIDI plugin to generate notes and some MIDI expressions. You can record this as a MIDI file or route its output within your host audio editor to a synthesizer plugin in it.

You can start with checking this Audio-to-MIDI plugins:

  • WIDI
  • Midifier (There are audio examples at the end of the page)

Since these are plugins and you need also a synthesizer to playback the output MIDI file with custom sounds, you should run these plugins within a sequencer or audio editor software with VST plugin support.

  • thanks gun ... can you explain this a bit more ? .. since I'm referring to recording written melodies using real instruments.. but maybe you have some crazy idea I never heard about lol
    – coroneddu
    Commented Jan 14, 2015 at 11:18
  • @coroneddu I've edited the answer with a little more detail. I hope it helps. Commented Jan 14, 2015 at 14:15
  • thanks Gun! .. already knew such things :) my focus is to maintain the performance nuances while those simply transcribe midi notes in a pretty much limited and simplistic fashion .. I think it's too early for such plugins.. those technologies are still pretty much primitive :) .. but thanks anyway..
    – coroneddu
    Commented Jan 14, 2015 at 17:11
  • @coroneddu Yes we need to wait a lot for that. Unfortunately sound recognition is the worst area of Artificial Intelligence. It is possible to get lots of things from an image data but sound is still very complex since the data in time and frequency domain is fit into 1-D space together with lots of additional information like location and characteristics. At least we are very capable about that, thanks to our brain, we can make music:) Commented Jan 16, 2015 at 16:04
  • yeps ... tho I must add that my question was more related to say, crazy plugins, that do unusual processings.. say, granular, or some weird filtering plus weird delays .. it's also about how they sound, the overall quality .. say, Absynth Aetherizer (based on granular) does something "unusual" but it sounds like crap to my hears.. while, say, sample based wise, S-Layer from Twisted Tools, approaches samples in a different way and sounds great .. course I could create various combinations of plugins myself, which I constantly do but I'm sure there are more than I know out there ..
    – coroneddu
    Commented Jan 17, 2015 at 14:07

Antares Kantos did this.

However, it was released around 2002, and appears to be no longer available from Antares, so I'm not sure it is possible to obtain a legitimate copy these days. A brief search suggests that there are sites which claim that it is freeware now, and possibly offer it for download, but I cannot vouch for those sites.

The general, brief sales pitch is/was:

Kantos analyzes any pitched monophonic incoming audio and instantaneously extracts pitch, dynamics, harmonic content and formant characteristics. This information is then used to control the kantos sound engine.

From the review at http://www.soundonsound.com/sos/Dec02/articles/antareskantos.asp (emphasis added):

During my tests, I tried Kantos with male and female voices, guitar and wooden flute as sources, and aside from detecting the wrong octave for some of the female vocal notes, it performed pretty well. I was particularly impressed by the way guitar and flute could be turned into quite realistic-sounding bowed instruments — the articulation feature definitely helps here — but don't make the mistake of thinking you can throw any old audio at Kantos and then get great results without making some adjustments. The input detector thresholds are critical to good performance, as is the use of clean, monophonic material, but as stated earlier, the chaotic results obtained when feeding in drum loops can be quite interesting. I also liked some of the effects that could be achieved by layering an untreated guitar sound with the synthesized output for an unusual double-tracking effect. Even the basic synth waveforms produce plenty of variety, so experimenting with the couple of dozen or so included wavetables widens the scope enormously, and if you get bored of those, there are more on the installation CD, and more still on the Antares web site.


It looks like Native-Instrument's The Mouth does something along these lines, too:

Designed to generate melodies and harmonies out of whatever audio material you feed it, THE MOUTH is a unique and exciting way to get creative results on the fly.

Sing, beatbox or send a drum loop into it and THE MOUTH comes to life as a tweakable synthesizer, bass, vocoder or talkbox-like effect. A playful interface provides you with plenty of room to shape your results as you go.

  • thanks for the input. and yep .. out of service :) I'd been curious to try it out
    – coroneddu
    Commented Jan 29, 2015 at 1:05
  • @coroneddu Alien Solo says it is a "poor man's Kantos / Vokator" and Super-Eel says it does "Audio In » Synthesized Sound Out like Kantos, Alien Solo" so they may be worth looking into.
    – D.G.
    Commented Jan 30, 2015 at 6:19
  • @coroneddu If you're still interested, it looks like NI's The mouth also does something along these lines (and is currently on sale), see my updated answer.
    – D.G.
    Commented Dec 1, 2015 at 18:12
  • thanks D.G. .. knew The Mouth since a while but is not what I'm looking for :) ..all the plugs I've found along the way sound too much "electronic" and the acoustic quality gets lost ... I've found that simply preparing and recording real instruments is my way to go for now .. I keep myself posted but nothing really interesting appeared on my radars so far ... but thanks for the update!
    – coroneddu
    Commented Dec 2, 2015 at 18:19

Okay, so time based effects like Chorus/Flangers/etc. are not allowed, but how about something like a ring-modulator? You could drone a synth note (and maybe play it with a foot pedal), and then modulate the synth with a ring modulator, or maybe switch it the other way around.

Conversely, you could try playing around with different Distortion/Saturation effects.

Another option is to vocode. Again, you're going to need some kind of underlying carrier, so again a synth-drone might work.

Heck, maybe you use an Audio to Midi plugin to get the general pitch of the note your playing, and then vocode the heck out of it. That would sound pretty sweet actually.

  • Thanks Jonathan ... in all my experiments I simply don't think technology is mature enough or either my requests aren't feasible :) .. Audio to Midi is still pretty rough and not capable of detecting/preserving all those nuances .. which is the whole point for me .. but thanks for the update you too!
    – coroneddu
    Commented Dec 2, 2015 at 18:22
  • Yea, the audio->midi won't do anything for nuances. That's where the vocoder and/or ring modulator comes in. Commented Dec 2, 2015 at 22:39
  • thanks Jon but I'm not interested in the Vocoder sound and with the Ring Modulators - I've tried few - I've never managed to come out with something interesting since I'm not into drone sounds but rather melodic stuff .. say using the guitar's nuances and playing (bends etc) but make it sound like something different .. been researching for years now but nothing satisfying (and that doesn't sound so "electronic") came up :)
    – coroneddu
    Commented Dec 3, 2015 at 23:28

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.