luxcache
patreon
NANOSCOPICS IN DIGITAL AUDIO - Part 1: Creative Tonal Mangling & Generative Melodies with sv1
🕑 Added 2021-11-11 17:54:07 +0000 UTC
NANOSCOPICS IN DIGITAL AUDIO
Part 1: Creative Tonal Mangling & Generative Melodies with sv1
In this Lux Cache tutorial series, our residential sound design master sv1 dives into the tinkering of microscopic textures to create his rich, nuanced and infinitely detailed compositions. To premiere the sequel to 2020’s ‘Organic Sound Design Masterclass’, sv1 explores his fresh new range of obscure and creative production techniques - investigating what’s possible spectral and generative effects inside Ableton Live and beyond.
This tutorial is available as both a Patreon text post and .pdf document format. We ask you kindly to not share Lux Cache content outside of the Patreon, our contributors rely on your donations.
~
CONTENTS
Here is a list of concepts and tools we will go over in this lesson:
- The Blank Slate
- Harmony to Midi
- Obscuring and Mangling Tonal Audio / Sample Chopping
- Pitch Hack
- Volume Gate + Pitch Hack
- Lossy
- Spectral Effects
- Spectral Blur
INTRO
Coming up with melodies and chords has always been the hardest part of writing music for me. I’ve always shied away from writing melodies and chords off of nothing, and over the years I’ve worked on gaming the whole process so that ableton will write things for me, such that I only have to go back and clean up afterwards. Some of my processes will require using third party vsts, but I’ll also try to suggest some alternatives so that you can recreate some of these processes elsewhere.
THE BLANK SLATE
(back here again…)
The hardest part of the song writing process for me is figuring out melodies and harmonies. I’ve never been good at just writing chords or melodies down without having some sort of base to come up with them. I also just have a really hard time dealing with the ‘blank slate’. So instead of trying to build it up from scratch, I prefer loading up familiar sounds such as samples or stems, and trying to work from there. I find it really handy to stem out old songs and keep them in my samples folder so that I can use old ideas that are mine and re-contextualize them, often drawing out new emotions and moods.
AUDIO TO MIDI
One of my favorite ways to come up with midi and tonal content for me to work with, is to take a song I’ve been listening to a lot, or old stems I’ve rendered out, and run them through Ableton’s native ‘Audio-to-Harmony’ algorithm.
Before I do that though, I’ll take whatever I want to convert into midi and run it through this external app called ‘key finder’ (linked below), that way I know what key the midi generated is supposed to be in so I can clean it up later. Once run through this audio-to-midi algorithm, most of the midi generated will be in key, but there will also be a lot of out of key notes as a result of resonances and other content the algorithm picks up one. Since we know what key it’s in, we can run that through the scale device to quantize all of that midi to the key we want to work in. If you don’t want to use the scale device and want to manually quantize all of the generated midi, open up Ableton’s piano roll-key feature, and push every out of key note down a step. This is how Ableton’s scaling device would do it.
(example output midi, notice how much of it is not in key)
**example audio - starting audio, 77 D**
**example audio - audio to midi, scaled, 77 D**
As you can see we’re starting to get something usable. Each bit of audio is different, sometimes you may get something very similar to what you started with and depending on the initial content you might not want to use that.
Very often I will go and clean up this generated midi and try to pull out manually ideas that stick out, but sometimes it’s fun to just feed instruments the mess that the algorithm picks up (especially using monophonic synths).
OBSCURING AND MANGLING TONAL AUDIO / CHOPPING SAMPLES
This method for getting tonal content for a song is super powerful, however we don’t want to go around stealing other people’s chords and melodies. So to prevent from arriving at something too similar (‘similar’, which I suppose is up to you, but I prefer to get really far away from the original sample I started with) I like to either effect the sample I want to extract midi from, or effect the resultant audio-to-midi sample. Here are some ways I like to accomplish this.
PITCH HACK
This is one of my favorite native effects in Ableton. It works very similarly to Half-Time or Gross Beat. All it does is reverse or re-pitch incoming audio. I like to turn up the reverse knob to 100% and then turn the d/w to around 50%, the result usually sounds like a mixed up and blended version of the original, and is usually dense enough to inspire some new ideas and ways to reharmonize.
Pitch hack is pretty buggy though, so usually I try to bounce out whatever it is that sounds good as soon as I arrive at it so that I don’t lose it. Often the result will not sound the same if you freeze/flatten the channel, so I recommend manually recording out whatever track you have it on. Also if you deactivate pitch hack and reactivate it, it will push your sample up 1 semitone, so bump it up and down one to reset the pitch of the output.
Below is what we left off with, but now with pitch hack applied.
**example audio - pitch hack, 77 D**
VOLUME GATE + PITCH HACK
Another way I like to chop up samples is to use some sort of gate to cut on every quarter or eighth, and then apply pitch hack after. There’s an ableton preset called ‘trance pad gate’ that I like using, but I gutted everything from it and just kept the utility and shaper on it, though the original version does produce some good results as well
Below is a different sample followed by the result of the gate and pitch hack being applied.
**example audio - Starting Audio, 77 D**
**example audio 5 - Gate + Pitch Hack, 77 D**
LOSSY
Next I want to talk about a couple of different plugins that all do similar things. Below I’ve linked a device that I’ve written in max, as well as some third party alternatives, that acts as a spectral gate, which produces that mp3 compression lossy sound (though what is compression, if not gating out data,in our case frequencies, to reduce file size). I really like using this to obscure sounds in order to discover new melodies and harmonies, and I’m also just a fan of the brittle mp3-like compression sound.
**One thing to note about this m4l is that it does introduce some latency, so I recommend setting your track delay to around -60 ms
Alternatives to this include Goodhertz Lossy Effect, as well as Reason Studio’s Audiomatic device (set to mp3).
MICHAEL NORRIS SPECTRAL AUDIO UNITS (OSX ONLY)
These devices are the main reason why I prefer working on music on mac. This set of devices comes with a million different effects, but the main one I use is called ‘spectral weave’. I’m not too sure how it works under the hood, but I know that if I feed it a sample that’s tonally dense, it’ll spit out these really gorgeous pads. Though I do encourage you to experiment with the different settings of course.
** Generally I don’t mess with too many of the knobs on this device, but I do almost always bring down the threshold, audio doesn’t get picked up for some reason unless you bring this down. **
Here I’ve used (**example audio - audio to midi, scaled, 77 D**) and below is the pad that spectral weave has spit out.
**example audio - spectral weave, 77 D**
Spectral weave is pretty buggy I’ve found, especially if you turn it on and off, it’s likely to crash your session. Again like with pitch hack, I prefer to bounce out everything from it immediately, and remove the effect as quickly as I can.
SPECTRAL BLUR
While I do prefer using spectral weave, Ableton’s native spectral blur does a lot of the same things, sometimes just as good (also with less bugs, and less latency).
You can also use this device to produce similar textures to the lossy effect because this also does act as a spectral gate. The residual knob lets in the dry audio, and the halo knob creates a reverb tale. More details on how to use this can be found in the Ableton Info panel.
Again I’ve ran this effect through our initial (**example audio - audio to midi, scaled, 77 D**) and the results are quite good, most due to how harmonically rich our starting point was.
**example audio - Spectral Blur, 77 D**
….But, if we run this once again through harmony to midi and scale it, we can get a result that’s further from our starting point, and invokes a different emotion with residual artifacts of the original.
**example audio - Spectral Blur, Audio to Midi, 77 D**
Also note that I’ve been using the default Harmony to Midi Instrument that is loaded up whenever this algorithm is executed. I usually detune the default a bit, and turn it all the way to synth, as well as turning the brightness all the way down.
CONCLUSION
The main point I want to drive home is that I like to mangle up old stems or samples using these effects so that I can infer new harmonies in order to write new music from that. This is a rather roundabout way to do this and I’d probably have an easier time just learning music theory and tinkering away at the piano at this rate. Though, I find it really fun trying to reharmonize old ideas or digging through samples in order to find something that inspires me. So many good and original ideas come from trying to ‘remix’ other songs or from iterating off old stems.
Links/Resources:
Keyfinder - http://www.ibrahimshaath.co.uk/keyfinder/
Michael Norris Spectral Audio Units - https://www.michaelnorris.info/software/soundmagic-spectral
Lossy
M4L Lossy - https://github.com/00ff1a/sv1-m4l
Based on: https://www.youtube.com/watch?v=69A1kGNFYIc&t=4s
Goodhertz Lossy - https://goodhertz.co/lossy/
Reason Studios Audiomatic - https://www.reasonstudios.com/en/reason/effects/audiomatic
~
sv1 is a texas-based musician and sound artist, his debut album, health, is available on all streaming platforms.
You can follow him on Twitter @sv1___ and Instagram @sv1.earth
