This week’s Disquiet Junto challenge: Make a piece of music by erasing aspects of a pre-existing track. Remove material from the earlier track using whatever techniques “erasure” could entail.
I chose the music I created for Disquiet Junto 0327 a few weeks ago called Deep Thought. That piece involved creating three lines of music: one in 3/4 time, one in 2/4, and one in 7/4, played simultaneously.
Here’s the music I created for this week’s challenge, which I call Shallow Thought:
And here’s the original, unmodified piece, in case you want to listen to it for comparison:
One significant constraint of this assignment is that the new piece must be based on the final audio recording of the old piece. I generally compose my music using Ableton Live, digital audio workstation software that lets you work with multiple tracks of MIDI (note) data and digital audio. I create separate MIDI tracks for each voice, edit individual notes, select instruments, apply effects to the instruments, and mix the tracks to create the final audio recording. At any time I can go back and tweak something in one of the tracks without having to re-record the other tracks. Even though I still have all the individual tracks of the original piece, they were not permitted to be used to create the new piece. This meant I had to dive into audio processing, something I haven’t done much of to date.
Comparing this to the world of Photoshop, it’s like making a new piece of digital art from an existing layered composition by editing a flattened version of the old piece, without being allowed to use the original layers.
I applied several kinds of processing to the audio of Deep Thought to create this new piece, Shallow Thought. I used several spectral filters to subtract different ranges of frequencies to create four tracks containing different spectral subsets of the original music: the high frequencies in one track, the mid frequencies in another track, and two tracks containing slightly different bands of the low frequencies. I used a Max for Live filter called Max CutHacker, which randomly cuts out pieces of the audio in time with the beat, applying it to three of the four tracks using different parameters. The volume of each track varies throughout the piece, removing different portions of the spectrum at different times.
One of the more interesting things to emerge was a percussion track that did not exist in the original music. The percussion sounds were an effect of applying the Max CutHacker filter on a very low frequency track where the pitched frequencies were almost totally removed.
I’m not sure this music is really “listenable” except in the context of comparing it to the original piece. It works in small doses, but two and a half minutes is a bit much for my ears. It was an interesting exercise and I will surely use the techniques I learned here in future music, but perhaps a little more selectively.