DSP algorithms

[blümchen]

Transients are very difficult to remain. That is in fact *the big problem* with all pitch shifters (time and frequency domain).

But I will try to find some Guitars.

Here are at least some other ones (Rhodes).
The same as the example above, but with polyphonic material:

http://test.dizainer.net/uniclono3.mp3
http://test.dizainer.net/uniclono4.mp3

.

[Urs]

The noise sounds a bit like time domain aliasing... maybe a problem with the overlap add?!?

Urs

[blümchen]

The noise sounds a bit like time domain aliasing... maybe a problem with the overlap add?!?

Urs

Yes. There is no good interpolation in the delayline. Therefore the noise. I'll try to get ride of that...

Here is the Guitar: (5 cent detuning 8 voices Uniclono. Switch between original and processed.)

http://test.dizainer.net/uniclono5.mp3

If you ask me. it sounds not very different to a chorus ... With all those many pitch shifters ...

.

[jmh]

Gotta say so far it sounds nice, despite not sounding that different to a chorus.

You are onto a good plugin here, man

Regards,

JMH

[Borogove]

... [the] dry signal plus delayed signal is a comb filter

a delayed signal plus a dry signal does not imply a comb filter

If the two signals are each arbitrary then I suppose you are right. If they are the same signal with one of the two being a delayed representation the other without any non-unity gain coefficients applied to either of them then the output will be the classic comb response. I believe this is what Borogove was trying to say.

Um, yeah. Wasn't trying to suggest that mixing dry Hendrix with delayed Sibelius would be a comb filter.

birrbits, consider a signal mixed with a copy of itself delayed by N samples. Frequencies which are multiples of Fs/N will be reinforced strongly by the delayed copy, frequencies which fall exactly in between those reinforcements will be cancelled. Comb filter. Here's JOS's overly-mathematical explanation.

Oh. Right. The feedforward comb filter has broad peaks and narrow notches in its amplitude response; the feedback version has narrow peaks. If you restrict your definition of the comb filter to the feedback version, then of course my description of the feedforward version doesn't apply. And pustekuchen probably defines a comb filter as the process of putting a hair-comb in an acoustic path.

[Urs]

Hmmm... yeah... pretty chorusish...

it would be interesting if there was a way to achieve an absolute frequency detune... so that you don't have cents but hertz. Kinda like dividing the scaling factor by bin order... if you know what I mean...

Cheers,

Urs

[Urs]

Another idea... what about leaving the lower frequencies undetuned... might be better for stereo processing...

Urs

[blümchen]

Gotta say so far it sounds nice, despite not sounding that different to a chorus.

You are onto a good plugin here, man

Regards,

JMH

Here's something, that you probably could find "orgasmic".

I really did not expect, that that kinda "multi pitch shifter" could do the following :

original: http://test.dizainer.net/analog_thin.mp3
fattened: http://test.dizainer.net/analog_fat.mp3

Can one do that with a chorus?
"It turns your monophonic thiny synthesizer into an omnipotent screaming *Unisono* monster machine!" ...


No. That is not real Unisono!


.

[birrbits]

... [the] dry signal plus delayed signal is a comb filter

a delayed signal plus a dry signal does not imply a comb filter

If the two signals are each arbitrary then I suppose you are right. If they are the same signal with one of the two being a delayed representation the other without any non-unity gain coefficients applied to either of them then the output will be the classic comb response. I believe this is what Borogove was trying to say.

Um, yeah. Wasn't trying to suggest that mixing dry Hendrix with delayed Sibelius would be a comb filter.

birrbits, consider a signal mixed with a copy of itself delayed by N samples. Frequencies which are multiples of Fs/N will be reinforced strongly by the delayed copy, frequencies which fall exactly in between those reinforcements will be cancelled. Comb filter. Here's JOS's overly-mathematical explanation.

Oh. Right. The feedforward comb filter has broad peaks and narrow notches in its amplitude response; the feedback version has narrow peaks. If you restrict your definition of the comb filter to the feedback version, then of course my description of the feedforward version doesn't apply. And pustekuchen probably defines a comb filter as the process of putting a hair-comb in an acoustic path.

Wasn't suggesting the Hendrix/Sibelius combo, but it might be interesting
pj stresses the right point (that I was hinting at) "without any non-unity gain coefficients" is the key, and of course feedback designs change things as well.
Again I was trying to be helpful... I can see that the help wasnt necessary...
pustekuchen would most likely insist that it is beneath him to tell you where his hairbrush is hiding.
Here's a fun thought (an evil prof threw at me once), sampling is just applying a comb filter to a frequency domain signal (treated as a time domain signal).

[jmh]

Here's something, that you probably could find "orgasmic".

I really did not expect, that that kinda "multi pitch shifter" could do the following :

original: http://test.dizainer.net/analog_thin.mp3
fattened: http://test.dizainer.net/analog_fat.mp3

Can one do that with a chorus?
"It turns your monophonic thiny synthesizer into an omnipotent screaming *Unisono* monster machine!" ...


No. That is not real Unisono!


.

You just made me ruin my last pair of underpants!!

Sounds great, some of the synth sounds... HUH



This really sounds sweet to my ears as I'm listening it on my headphones. I can already think of a couple of my tracks that would really benefit from this

Regards,

JMH

[G-Spot]

Digital "unison" is anything out of some digital processing that sounds like actual unison but it isn't.
An "unison effect" of the type discussed in this thread is a somewhat magic box that sounds like unison but works on a "solo" audio input, as opposed to a trivial mixer of ready-made synchronized different signals. A "chorus" effect is an unison effect implemented in a specific way that someone finds cheap.

There *is no* such categorizing of "analog" or "digital" unisono. And a chorus is even quite different.

Analog? This is the KVR forum, I didn't even think about analog signal processing.
For the purpose of this discussion, all analog and digital effects fall into the same category of "fake unison" and two analog or digital synths played by two people count as real instruments exactly like two pianos.
I was comparing actual instruments with signal processing tricks; of course there are border cases like organs, harpsichords and many synthesizers that can mix different sounds by themselves.

[helium]

@pustekuchen:

Nonsense. Then you actually have an effect (a reverb effect regarding your description). And nothing else.
A reverb is in fact an effect, which can be applied afterwards and to polyphonic material and whatsoever...


A synthesizer plays Unisono only then, if it playes the voices together in real Unisono mode. Like the different instruments in the orchestra, which play "Unisono" regarding the partiture.

But that synthesizer playes simply a chorus effect, if that effect is applied afterwards. Then it isn't Unisono anymore.

We discussed that already excessively.

There is also a contradiction in your thesis: You talk about "equal pitch" and the at same time about "detuning". So what?

I mean, begin to read all the thread from beginning, because what you have said is all quite clear to us at this moment ...

In meiner Antwort ging es nicht im geringsten darum, ob die einzelnen Stimmen jetzt verstimmt sind oder nicht. Und einen Chorus habe nicht nichtmals angedeutet geschweige denn erwähnt.
Du hast die ganze zeit darauf bestanden, dass wenn ein Synthy Unisono spielt alles komplett mehrfach ausgelegt sein müsste ich dagegen bin der Meinung, das es genügt lediglich die eigentliche Klangerzeugung mehrfach auszulegen.

[blümchen]

In einem Synthesizer ist so ziemlich alles "Klangerzeugung". Es sei denn Du hast noch ne Kaffeemaschine dran ...

[blümchen]

The noise sounds a bit like time domain aliasing... maybe a problem with the overlap add?!?

Urs

I searched like crazy the bug inside my interpolation code, doing up to step-wise debugging ...

But it was finally simply a wrong (integer<->float) conversion.

.

[blümchen]

Hmmm... yeah... pretty chorusish...

it would be interesting if there was a way to achieve an absolute frequency detune... so that you don't have cents but hertz. Kinda like dividing the scaling factor by bin order... if you know what I mean...

Cheers,

Urs

No. Not exactly knowing, what you mean.
Dosen't that need a pitch detection then?

I mean, a delay line pitch shifter does not do any frequency analysis, but simply interpolation on base of a playback rate ratio.

.