thor: smoothing audio to cv gate roughness?

This forum is for discussing Reason. Questions, answers, ideas, and opinions... all apply.
Post Reply
owlis
Posts: 41
Joined: 21 Jul 2019

23 Jul 2019

Hi I am trying to make some organic rhythmically evolving trance gates in thor
I am patching like this in thor’s matrix:
AudioIn 1 to Cv out1

then on the back I plug Cv out1 into Cv in1

and use that to drive the frequency or drive to create a gated effect
however the sound isn’t smooth it is buzzy and granular
Is there a plugin that will smooth that out or a better way to wire it?

Cheers
Pete

rhysonic.com

User avatar
Loque
Moderator
Posts: 11187
Joined: 28 Dec 2015

23 Jul 2019

You are trying to modulate a audio signal based on the frequency of an audio signal, which is reduced to CV resolution. Ofc this sounds like crap in most cases. If you want to modulate, than use audio rate modulation if possible, that means modulate the Aduio signal directly with the filter or just use LFOs, which sound smoother.

But AFAIU you do not want to use audio modulation, you want a rhythmic gating effect. You need a signal, that is slow like 1/32 step which triggers something, which makes the audio quiet, like volume or closing a filter.

You can use Thor's sequencer to trigger the Filter Frequency or Volume. Exactly for things like this Thor's sequencer has additional CV signals. Use the Seqeuncer CV signal as a Source in the Modulation Matrix, adjust the Amount, set the Target e.g. Filter Frequency, and tha run the Sequencer. Note, that you need to provide values for CV Steps in the Sequencer.
Reason12, Win10

owlis
Posts: 41
Joined: 21 Jul 2019

23 Jul 2019

Hi Loque thanks for your response. As far as using the thor sequencer cv to control the filter frequency etc... I have been doing that quite a bit already. What I am imagining doing is taking a rather complex tabla performance, for example, and somehow trying to use that very complex syncopated poly rhythm as a gate for a pad sound somehow. Since an audio waveform has a visual representation where you can see peaks and troughs, I would imaging there is a way at least in theory to covert the waveform to a simplified bezier representation and then possible further modify it to be a square wave at each peak that it finds.

The closest example of what I am talking about is a function in a program called After Effects that can convert a waveform into a simplified vector representation. Like in the image in this link

https://macprovid.vo.llnwd.net/o43/hub/ ... tep-2b.png

I would also be happy to just use some rack extension (if one exists, that can convert audio into smoother cv data)

Cheers
Pete

User avatar
Loque
Moderator
Posts: 11187
Joined: 28 Dec 2015

23 Jul 2019

Ohhhhkayyy...That is something different again. You want to map the volume curve of a sample to another instrument.

You can do this with a Compressor scanning the audio signal and creating the CV signal for a volume change. To do this
* If you want to use the original audio signal, drag a Audio Splitter below your audio signal, connect a Output of the Splitter to the Mixer or Return, depending on your signal flow
* Drag a MClass Compressor below your Splitter and connect another output from the Splitter to the Audio In of the Compressor
* Flip the rack to the back and check the CV output from the MClass Compressor
* Connect https://www.propellerheads.com/shop/rac ... -analyzer/ to this CV out if you like - its just for visualization and a few other cool features
* Connect the CV signal to Thor as you already did and route it to your Filter
* Now playback and adjust the Threashold and Ratio of the Compressor until you can see the Curve Moving in CVA-7. Adjust all values to taste
* Note, that the CV output from CMlass Compressor is inverse, so you might flip the sginal in Thors Mod Matrix to negative/positive

Btw, this is called "ducking".

Hope that makes sense...
Reason12, Win10

owlis
Posts: 41
Joined: 21 Jul 2019

23 Jul 2019

Thanks Loque I think that is what I am looking for :) Will try it when I get home.

Thanks
Pete

User avatar
selig
RE Developer
Posts: 11739
Joined: 15 Jan 2015
Location: The NorthWoods, CT, USA

23 Jul 2019

Sounds like you're looking for an "envelope follower" (to convert audio to CV), connected to a filter which creates an "auto-wha" or "filter-gate" effect in this configuration.

My approach:
MClass makes a great enveloper follower: Simply set Threshold to the lowest, Ratio to the highest, and adjust Attack/Release to taste. Attack/Release will control "smoothness" of the resultant control signal/CV.
Then use modulation depth at the filter end to control the amount of the effect rather than ratio/threshold. Then route the MCLass CV out to any CV in on a filter, or connect to a gain stage (VCA in Thor) to control level.

With apologies in advance to Logue, this setup would generally be called side-chain gating or dynamic filtering. "Ducking" is a specific (and different) type of side-chain setup where you use an enveloper follower to REDUCE the VOLUME of one track in response to a different track, and was originally used in advertising to automatically reduce the level of the music bed when the announcer spoke. In this case (above), you would be doing the opposite: INCREASING the filter or audio level in response to another audio signal.
Selig Audio, LLC

User avatar
Reasonable man
Posts: 589
Joined: 14 Jul 2016

23 Jul 2019

Fo rsmoothing of all cv signals in the past i 've taken the cv signal into pulveriser's enveveloper followe's input at the back ...tweaked the attack ,release settings and taken it back out the follower's output at the back. There's a propellerhead video on this ..but Selig's method looks ideal....also Poobear did an excelent video on envelope followers on his channel recently with a downloadable combi

owlis
Posts: 41
Joined: 21 Jul 2019

23 Jul 2019

Thanks Selig and Reasonable Man :) I will try that envelope follower technique.

Cheers
Pete

rhysonic.com

Post Reply
  • Information
  • Who is online

    Users browsing this forum: stm.mitchell and 20 guests