Replication is often how we gain skills; it’s how we build our knowledge-base as a foundation to build new things. Singing intervals trains your voice and ears so you can have the tools to put those intervals in the context of a piece of music. Drawing an object in front of you teaches you to see depth, lines, light, and colours more clearly, so you can take that knowledge and create your own art. Writing poems in various forms gives us a structured way to experiment and find our own voice. Attempting to recreate sounds via subtractive synthesis is teaching me (slowly) how to listen for overtones rather than letting my ears summarize their effect down to a vague term for the colour/timbre of the sound.
It is so easy to get lost in the shifting sounds as I make adjustments with various knobs, so it’s very difficult to accomplish anything that I initially set out to do. However, the fact that I can get so lost in experimenting tells me I must be learning something, and learning is my general mission, right? It’s like if someone sets out to draw a hand, and they spend all their time following the lines within the hand, and learn about the texture and depth created by those lines, but they beat themselves up over not actually finishing the drawing of a hand, even though they still learned so much from the work they did. So, I guess what I’m trying to say is that I’ve been learning SO MUCH STUFF, but I couldn’t figure out how to package it up nicely in a post, so I’ve been feeling like a horrible student, and I’ve finally realized how silly that is. Learning new things is a messy process, and presenting it as a polished portfolio-ready product can inhibit the actual learning.
When I was messing around on the Roland GAIA, initially trying to make a clarinet sound, I decided to try using a noise oscillator to experiment with filters, to listen for what happens to the overtones. When I left the building after that, I could identify the different layers of noise from various buildings and machines on campus. IT SOUNDED LIKE A SYMPHONY. That was one of my biggest learning experiences so far, and it came from taking a detour from what I had set out to do. That moment when I realized how much I learned from a detour, was when I realized how silly I’ve been for worrying about following my plan. Plans shouldn’t detract from the learning that happens along the way.
The fact is, I don’t have time to cover everything I need to cover, internalize it, and present my findings in a neat little package. So, I’m going to treat this blog as a documentation of my learning, rather than a perfectly organized presentation of topics. There are tons of articles of wonderfully organized teachings of topics, and before I can add to those, I have many things to learn. Perhaps one day, I’ll be able to assemble my ideas about electronic music into some nicely organized doctoral thesis, but for now, I just need to learn as much as possible. So far, this course is teaching me more about learning it is about anything I’m trying to learn, and I think that’s pretty cool!
I could muse over ideas about learning for hours, but I think I’ve said most of what’s been on my mind, so let’s jump into the little bit I initially set out to present and get it over with so I can move on to learning and experimenting with the next section I’m going to cover.
A vaguely structured approach
I got fairly behind in my schedule that I had initially envisioned, due to life, learning, and finding balance. In order to push through, I decided to combine two of my sections into one: Sound and Synthesis Fundamentals, and Synthesis Concepts. Sound and Synthesis Fundamentals is a section where I’m supposed to study acoustic sounds via oscilloscope, and make an attempt at recreating those sounds, and present my findings via comparison. That section was meant to reinforce the idea that synthesis is something different from acoustic instruments, and they should be seen as separate, not replacements for each other. With so many sonic possibilities, why replicate sounds that can already be made by existing instruments? Synthesis Concepts was meant to cover subtractive synthesis, FM synthesis, and granular, wavetable synthesis.
I learn concepts best when I have a practical application, so I set out to study these concepts by using them to create a clarinet sound after studying recordings of clarinetists via oscilloscope. I thought I would package up all my findings in one tidy blog post, but that idea wasn’t realistic, so I’m going to post about each synthesis concept separately, and I will not restrict myself to replicating a clarinet sound with the next two posts, because I think I got what I need from that exercise with this experiment.
Studying the sound
I routed the audio from a few recordings through an oscilloscope to study what clarinet sounds look like. I recorded the oscilloscope with my phone, and since I can’t upload videos here unless I upgrade to premium, here are some images from that video:
I could have uploaded the video to some other platform and linked to it, but I think these screenshots are enough for the purpose of this blog.
As you can see, the sound isn’t exactly consistent due to the colouristic possibilities within the clarinet sound, so I knew that I would have to arbitrarily choose which clarinet sound to synthesize, and I chose to make something like the bottom left corner picture.
I found some sources to learn about subtractive synthesis:
And I used the information from these to carve out a sound on the Roland GAIA SH-01 that could look like a clarinet sound.
Synthesizing the sound
I started with a square wave, and I wanted to make it darker and more focused, so I used a low-pass filter. I messed around with the levels of the cutoff and resonance until I found something that looked like the sound wave I was trying to replicate. I got distracted many times by the different sounds I could make, but eventually I settled on something, because I knew I wouldn’t be able to make a sound that satisfies my clarinetist ears. I couldn’t figure out exactly what was missing, but the closest explanation that came to mind was the intensity and direction that I’ve always been taught to strive for in my sound. Whether it’s my lack of experience, or the inherent limitations of synthesis, I was incapable of creating a sound that even comes close to what my clarinetist ears look for–which might be the reason it’s taken me a week to put this post together.
I forgot to get pictures of the oscilloscope, but I’ll get those when I’m in my lab this week.
And here they are:
I did get this recording, since I was using Stravinsky’s Three Pieces as a way to study clarinet sound waves, I decided I would make an attempt at playing the opening of that with my robotic clarinet-ish sound:
(Hey look! Embedding things instead of linking to them. Making progress…blogress?)
In attempting this, I also learned that it is much easier to play this piece on a clarinet than on a keyboard, given the leaps involved.
General thoughts about this exercise
I definitely learned a lot. Most of what I learned was not directly related to my assignment, but I still learned it through this assignment, and that added value to the experience. I definitely want to just get this posted so I can move on though; I have so much to do, and this is only the beginning!
Next up: exploring FM synthesis via FM8, which will probably result in a shorter post than this one simply for practicality of getting through this course. I can explore topics in greater detail when I don’t have a time limit, but I’d like to avoid running this into the school year, since that will be stressful enough as it is.