Creating the Shepard Tone Illusion with any sound effect, in Kyma

Question

Hi everyone,

I came across with this video that shows a way (starting on 1:07 in the video) of creating the Shepard Tone illusion, by basically using any sound effect in Pro Tools. In my search of finding a way of achieving the same effect in Kyma, I found Cross input w/Endlessly rising Shepard tone thru Vocoder  and also SheppardGlissando and ShepardContinuousSpectrum prototypes. But, I can't seem to achieve the same results as the ones shown in the video. 

1.-  I used Cross input w/Endlessly rising Shepard tone thru Vocoder for this sound (sfx used in Sample Up/Down Prototypes).
 

a) Why is there only a frequency tone generated, that goes up in pitch (that's actually the Shepard Tone effect),  instead of just geting only the SFX (used in the SampleDown and SampleUp Prototypes), going up in pitch? 

b) What am I doing wrong here?

c) Is there a way to just lose the frequency tone?

2.- It seems that the SFX in the Sample Up/Down is used only as a "modulator", or "trigger", for the upper part of the Sound. ( Starting from the Resynth w /saturation Protype all the way back to the Ramp Sample Prototype).

Am i wrong in this assessment?

3.- In the SampleUp/Down Prototypes, the 1 repeatingRamp: !Dur s and 1 - (1 repeatingRamp: !Dur s) expressions were used in the Scale parameter, in order to simulate the fades in and out, demonstrated in the video.

Was that a correct decision to control the Scale parameter, or would you recomend something else, in order to achieve the fade in/out process, explained in the video?

4.- In the SimplePitchShifterUpwards/Downwards  the following expressions were used in the Interval parameter, in order to simulate the pitch up and down, demonstrated in the video:
(1 repeatingRamp: !Dur s) * !IntervalUpwards and (1 - (1 repeatingRamp: !Dur s)) * !IntervalDownwards respectively.

Was that a correct approach to control the Interval parameter or would you recomend something else, in order to achieve the pitch up/down process, explained in the video?

5.- Is there any other way/technique or Sound that you would recomend me to look into it?

Thank you very much for your help and time!

Best,

Marco 

Answer 1

Some other approaches, in the Sound Browser, double-click Kyma Sound Library, then Search for Sound name = shepard. Lots of synthesized examples come up but the example called 'Disturbing (endless Shepard FilterBank)' is another one that can be applied to any input sample. It's close to what you were doing with the ShepardContinuousSpectrum but instead of using the spectrum source to control oscillators, it uses the shepard spectrum source to control a bank of tuned bandpass filters.

Other examples that come up in that search are also kind of fun (including surround glissandos that make you feel like you are endlessly falling down the rabbit hole ;)

Answer 2

A couple of examples that might be good starting points:

Kyma Sound Library > FX > Whooshes hits bys.kym has two examples — Hell's Elevator and Hellish (more) elevator that are variations on Pete Johnston's rotary phaser concept.

There's another approach you could try in the Gallery:

• Select a sample in the Sound Browser

• Disclose the Gallery settings and check: Pitch, Players, Mangers, & Replicators (Uncheck Modulate parameter values).

• In Gallery > Manglers++ play Replicator DriftingSampleCloud++

• In the VCS: Direction = 0.25, IntervalSpread = 7.0, Octaves = 5, ReverbTime = 4

There are also other ways to approach this (to be continued...)

Comments

Thank you for your help and time, I will try your suggestions. Looking forward for reading about the other ways too!

---

After trying your suggested approach by using the Gallery method, I was wondering if there's any suggestion in using instead of a sample, a generic live, or an audio input prototype instead, so I could feed Pro Tools into Kyma with any fx?

Best,
Marco

I am using Dante Via/ Controller and the Dante USB 2Ch I/O adapter to feed Pro Tools into Kyma and viceversa.

---

Hello Marco,
If you drag and drop an AudioInput module onto the current input Sample module, that will apply the same processing to the live input as it was applying to that sample. You can specify which audio input channel in the Audio Input parameters. Does that give you the result you were looking for? Thanks!

---

Hi SSC,

Thanks for your suggestion. I did try the Generic Live and the AudioInput earlier, but unfortunately the resulting sound is static, because I am loosing some faders that the Morph1dSampleCloud module offers, like f.ex: density, direction, DurJitter, etc...  

Best,
Marco

Answer 3

Hi Marco,

To use the DriftingSampleCloud example from the Manglers Gallery with a live input, add a MemoryWriter to capture the live input into a buffer and change the Morph1DSampleCloud to point to that memory buffer. Here's an example of a live endless Shepard SampleCloud that you could use as a starting point. There are a lot of things you can adjust on it, like the CaptureDuration in the MemoryWriter, whether you want to capture the live input once or whether you want to set it to Cyclic, whether you want to capture new input only on a trigger or continuously, and so on.

Have fun! 

Comments

Thank you so much SSC!

Answer 4

Continuing SSC's answer, the Morph1dSampleCloud can also use MemoryWriter buffers. You can make a flow where the live input signal is fed into 1 or more MemoryWriters and their buffer names are put in the Morph1dSampleCloud instead of file names. Then just check the "FromMemoryWriter" parameter in the sample cloud. 

 

Like....

In this example I've got the audio input going in to two MemoryWriters each with their own record buttons so I can record into two buffers and then morph between them with the Cloud player. 

Here's the Sound

 

Comments

Thank you very much Alan! What a coincidence to be talking on our today's 1: 1 class about the memory writers:)