Random Walks in Sound
A random walk drifts by small random steps. Apply it to pitch and you get a wandering melody. Apply it to brightness and the timbre slowly evolves. The step size controls how wild it gets.
Walking Through Parameter Space
Flip a coin. Heads, go up one step. Tails, go down one step. That is a random walk. Over time you drift away from where you started, but each individual step is small and predictable. The result is smooth motion without a plan.
Apply this to a synthesizer and you get parameters that evolve on their own. Pitch walks up and down in small intervals, creating melodies that wander without repeating. Filter cutoff drifts, so the timbre brightens and darkens on its own. Volume walks, giving the sound a breathing quality.
The step size slider controls how big each random step is. Small steps give gentle, barely-noticeable drift. Large steps make the walk erratic. The "Replay same walk" toggle locks in a seed so you can compare the exact same walk with different step sizes.
Try it -- press Play and watch the three walk lines scroll across the graph. Turn up Pitch Walk for a wandering melody. Turn up Filter Walk to hear the brightness drift. Toggle "Same Seed" on, then change step sizes to hear the same walk at different scales.
Why This Matters
Random walks are one of the simplest forms of generative music. A Turing machine sequencer (Part 93) uses a shift register for its randomness. Here you are using pure coin flips applied directly to synth parameters. The results sound organic because the motion is continuous. There are no sudden jumps, just gradual drift.
Brian Eno used similar techniques in his ambient work, letting slow random processes create music that evolves over hours without repeating. Modular synth patches often include sample-and-hold modules feeding random voltages to oscillators and filters. This is the same idea: let chance shape the sound, and set the boundaries so it always sounds good.