VST plug-in for envelope / LFO generation

Started by wookie, March 08, 2011, 11:38:08 AM

Previous topic - Next topic

wookie

Has anybody tried creating a VST plug-in to create an envelope detector output to drive off-board analog gear?  I was thinking of coding something up like that, but wanted to check if it had been done first. 

The idea I have is to make a VST plug-in that performs envelope detection in code, and output just that....no dry signal at all.  The idea being that I'll route the output to a dedicated hardware out and run it into some sort of signal conditioner to get the levels to where I need with the appropriate amount of drive current available on tap to avoid loading issues.  Then run the output of that to control various envelope controlled effects.  This will let me tweak the envelope detection algorithm, and do lots of other fun tempo related stuff, in the digital realm and keep an all analog signal path (not a huge deal to me...but the idea sounds fun). 

I can see track delay possibly being an issue with this idea, but it's sounds fun enough to try out.  The other would be the fact that the output is likely AC-coupled, which could be an issue for some types of sound sources (organ-ish).

wavley

Well, the MOTU Volta is kinda like that for CV, maybe you can look to that for inspiration.

http://www.motu.com/products/software/volta
New and exciting innovations in current technology!

Bone is in the fingers.

EccoHollow Art & Sound

eccohollow.bandcamp.com

wookie

Thanks for the link!

They do specify that you need a DC-coupled audio interface to make it work. I've got a Delta 44, and it doesn't list any coupling information...but does show a frequency response lower limit of around 22 Hz.  Probably won't work with my setup.  One of the cheep USB DAQ devices like NI's USB-6008 may do the trick, but I'm betting that getting the synchronization right would be a big pain.

CynicalMan

I think high-pass filtering on the output of the sound card would be the biggest issue. You could modulate the amplitude of a high-frequency wave with the envelop signal, which would require a circuit on the output to make it DC again. Something like: precision rectifier --> low-pass filter --> amplifier --> other circuits.

ExpAnonColin

Quote from: CynicalMan on March 08, 2011, 05:36:21 PM
I think high-pass filtering on the output of the sound card would be the biggest issue. You could modulate the amplitude of a high-frequency wave with the envelop signal, which would require a circuit on the output to make it DC again. Something like: precision rectifier --> low-pass filter --> amplifier --> other circuits.

I agree here.  I think an idea like this would work well with the Volta (or similar system) and a compatible interface, or something like above, but not a lot else (unless your interface has the ability to generate really high frequencies you'd have to worry a lot about interference).  Envelope detection is generally not that computationally intensive so you'd probably be better off just using a microcontroller.

wavley

New and exciting innovations in current technology!

Bone is in the fingers.

EccoHollow Art & Sound

eccohollow.bandcamp.com

wookie

What I had in mind was similar to developing a free-ware Volta in some respects.  I wasn't aware of that project before I posted.

I agree that a simple envelope detector on it's own is better off being done in HW, but I had some plans for bringing in host BPM information to bring it into some other directions.  I could go the route of a micro that detects MIDI clock, but I'm not sure that the project would find much of an audience that way.

I like the idea of de-modulating a signal back down to baseband to get a CV that way.  It's definately not the most practical, but it's got a whole lot of interesting possibilities in there.  And it sounds fun taboot!


This project likely doesn't have the legs I initially envisioned, but I'll keep everyone posted on anything I develop.  Thank to everyone for their input / help.

ExpAnonColin

Quote from: wookie on March 10, 2011, 04:23:35 PM
I could go the route of a micro that detects MIDI clock, but I'm not sure that the project would find much of an audience that way.

Not to play the devil's advocate, but I'd think a project which extracted tempo information would get MORE of an audience because you'd have the option of using it without a computer - ie, with hardware that transmits a MIDI clock.

-Colin