KIYMA
Key Features
- Captures the recent phrase and replays its steps in a new, tempo-locked order.
- Five tuned sweet spots move quickly from intact replay to dense, stretched, or heavily sliced results.
- DENSITY, REACH, and PATTERN shape how often steps jump, how far they travel, and which fixed map controls the chop.
- LENGTH can read half a bar through four bars while always replaying the result in one bar.
- EDGE and STEPS adjust cut softness and choose 8-, 16-, or 32-step resolution.
- Dry/wet and output controls make it possible to blend rhythmic disruption into a mix rather than fully replace the source.
- Uses past incoming audio for zero added latency and deterministic results at the same settings.
Description
KIYMA is a tempo-synced audio mincer from SubverseLab that captures the phrase immediately before the playhead, divides it into steps, and reorders those steps inside a bar. It turns a steady vocal, loop, synth, or drum part into repeats, cuts, octave-shifted compressions, and other rhythmic disruptions while remaining locked to the DAW timeline.
The five switchable sweet spots offer deliberately different starting characters: LEAN leaves the phrase intact, while MINCE, CHUNK, GRISTLE, and BONE move through denser slices, longer captured phrases, or a stretched half-bar. Raising a switch recalls its tuned settings, and changing a knob returns the effect to hands-on control.
DENSITY determines how often a step jumps, REACH limits how far it can move, and PATTERN selects one of the fixed chop maps. LENGTH sets the amount of recent audio KIYMA reads, while EDGE, STEPS, dry/wet, and output controls refine the cuts, resolution, balance, and level.
Because KIYMA reads audio that has already arrived, it reports zero added latency and does not rely on random playback decisions. Its fixed patterns make a given source and setting repeatable, making it especially practical for beat-synced fills and controlled glitch texture rather than one-off accidents.
Frequently Asked Questions
What does KIYMA do to incoming audio?
KIYMA keeps the last complete block of bars that reached the insert, cuts it into steps, and decides whether each replayed step stays in place or jumps to a nearby one. The selected capture length is always replayed in one bar, so longer captures can also change pitch and speed.
Are KIYMA's patterns random?
No. Its chop patterns are fixed tables, and the same settings on the same song produce the same result at any buffer size. The DENSITY control changes the chance of step jumps within that deterministic behavior.
Does KIYMA add latency?
SubverseLab states that KIYMA only reads audio that has already arrived and therefore adds zero latency. It follows the DAW tempo, bar position, and time signature for its timed playback.
How do I download KIYMA?
Start from the official SubverseLab KIYMA page and complete the free checkout with an email address; no card is required. The download is then available from the product page and the associated account page.
