All DSP nodes, A to Z
Every node on one page. To narrow the list by category, target or published examples, use Explore.
A
Exponential ADSR envelope state machine in VCA form over the shared one-pole coefficient/step and stage thresholds: gate-on is a full retrigger, gate-off releases from…
Exact whole-frame alignment without clipping, interpolation or feedback.
Equal-power sine auto-panner: one bounded f32 phase accumulator drives both channel gains with no per-sample allocation.
B
Band limited table in Oscillators
Band-limited wavetable playback: per-frame FFT into a sixteen-level half-octave pyramid (256 down to 1 harmonic, like PeriodicWave), Nyquist-safe level crossfading, cubic cycle…
Canonical f32 RBJ biquad state, coefficient design, tick and reset semantics for the seven kernel filter types.
Sample-rate and bit-depth reducer: anti-imaging one-pole, held-sample phase accumulator, quantization steps, boost and dry/wet with boost compensation.
Tam-tam nonlinear upward energy cascade as a per-mode arrival-time split: deferred excitation fraction, one-pole rise coefficient, envelope step and the audible per-mode gain…
Hall feedback-delay network over caller-owned delay storage.
Two-rotor speaker approximation with independent motor inertia and caller-owned Doppler rings.
Dispersive spring-style waveguide with coupled caller-owned delay loops.
Non-regenerating fixed delay: the feedback delay's sibling, writing through the same bounded store with the regeneration term exactly zero so all three delay behaviors share one…
C
Chebyshev first-kind polynomial waveshaper over the shared fixed-4x oversampled interpolation and FIR decimation stage.
Chord colour: up to two added tones from the key's scale over a chord (pitch-class mask and root), filtered by documented avoid-note rules and ranked by interval-class…
Chord voicing of a stacked chord of one to seven tones into MIDI notes: root position, voice-led (least motion from the previous chord), drop 2 and spread…
Chorus in Modulation
Mono chorus: a sine-modulated bounded fractional delay over the canonical fractional-delay line, with optional regeneration, a dry/wet mix and a degrees-authored static LFO…
Classic nonlinear amp v1 pre-cabinet voice: input high-pass, pre-emphasis, 4x oversampled preamp and power waveshapers around a three-band tone stack…
One prepared planar stereo clip at any 8-384 kHz source rate: sample-clock transport with run/loop/normalised seek, 4-point Hermite resampling to the host rate and a 5 ms…
CHOMPI TEMPO firmware port of the clocked glitch delay (granularDelay.h): tempo-synced stereo f32 delay whose repeats are re-voiced on clock edges by newlib rand() draws…
Original Mutable Clouds engine translation.
Wet hard clipping, asymmetric saturation and rectification with tone/DC filtering.
Deterministic coloured-noise source: seed policy, [-1,1) white mapping, pink/brown colour filters, fixed-crossover signed tone tilt and output level.
Bounded constant/DC source for modulation graphs and cross-lane parity probes.
Constant-power pan law over a bipolar -1..1 position returning UNROUNDED doubles (`cos`/`sin` of `(pan + 1) * PI/4`), for voices that fold the pan into a per-voice constant gain…
Control-to-gain mapping law (rectify, normalize, optional invert, power curve, bounded min/max range) and the VCA that applies it.
Bounded zero-latency direct FIR convolver capped at 256 mono taps with format-neutral impulse loading.
Cross feed in Delays
Canonical cross-feeding stereo delay: two fractional lines with per-line times, split same-side/cross-side feedback, in-loop high-pass/low-pass tone, quadrature read-position…
D
MIT DaisySP BLEP support helpers shared by the ported oscillators: the polynomial band-limited step and integrated-step residuals, the turn-based sine and the port's conditional…
True-stereo Dattorro network ported from the pinned Synthlet MIT/STK reference.
DaisySP DcBlock leaky differentiator with exact f32 semantics and a fixed per-sample 0.99 pole, as the CHOMPI firmware uses it.
Canonical f64 decibel-to-linear gain and RBJ amplitude conversions with ECMAScript edge behavior.
Wet-only DaisySP Decimator port: raw sample-and-hold cadence and defined fixed-point bit clearing without filtering or added latency.
Delay pitch shifter in Spectral
Dual sweeping-delay pitch shifter over caller-owned prepared stereo rings.
Direct overdrive in Distortion
Direct zero-latency DaisySP overdrive with bounded drive-dependent gain and soft clipping.
Analytic cardioid-family directivity (omni, figure-eight, hyper-, sub- and cardioid with the p parameter), angle functions, direction vectors, cartesian/spherical conversion…
CHOMPI TEMPO firmware port of the clock manager: two division lanes plus a fixed eighth-note lane over a free-running 24 ppqn timer or external MIDI clock…
CHOMPI firmware port of the one-knob DJ filter: smoothed resonant low-pass below centre and high-pass above, two multimode cascades in series per channel, with TAPE…
Drive in Distortion
Canonical fixed-4x rational-drive transfer, FIR state, tick and reset semantics.
Shared drum-kit model data and contracts for canonical percussion consumers.
E
Envelope-driven SVF composed from canonical low-pass and detector primitives.
Canonical peak/RMS envelope-follower state, branching one-pole attack/release detector, and the standalone mean-square window a separated compressor detector needs.
Three-band low-shelf/peaking-mid/high-shelf EQ composed from the canonical biquad unit with a fixed 0.7071 shelf Q.
F
Single-precision control slews rounded operation for operation: DaisySP fonepole and the stmlib SLOPE asymmetric rise/fall step, as the CHOMPI firmware smooths its controls.
Regenerating fixed feedback delay over the shared delay primitives, with a bounded active length inside a preallocated Float32Array line so the stored feedback quantizes to f32.
Felt hammer in Physical models
Unilateral nonlinear felt compression and hammer momentum against a passive impedance load.
Short sine-modulated fractional delay with signed feedback, phase-locked stereo support, and staged lifecycle clearing.
Float clamp in Math
NaN-ignoring IEEE fmin/fmax clamp with newlib signed-zero ties, as DaisySP fclamp computes it for the CHOMPI firmware.
Shared FM operator-routing algorithms and stable algorithm lookup.
Prepared operator-count-flexible phase-modulation graph with weighted source-to-target routes, deterministic one-sample feedback for cyclic connections, and independent carrier…
Four-operator phase-modulation keyboard voice: six curated algorithms with per-voice latched wiring, the stepped ratio table (harmonic, sqrt(2)-family inharmonic and five…
CHOMPI firmware port of the stmlib peak-follower Limiter: asymmetric slope follower and reciprocal gain into the rational soft limit, plus the firmware's compressor and hard paths.
DaisySP FormantOscillator port: a windowed-sine formant generator whose carrier reset is band-limited, with the phase shift applied as an explicit per-sample increment.
Fractional in Delays
Canonical bounded f32 linear-interpolation delay and feedback-delay state, read/write, tick and reset semantics.
Single-sideband frequency shifter: two fixed four-section Niemitalo allpass banks and phase-locked quadrature carriers.
G
Bounded linear output-gain stage as a per-channel double multiply over one shared parameter state.
Bounded decibel gain with a mute gate, using the canonical dB-to-linear conversion.
Struck-idiophone gamelan voice as a modal resonator bank over the shared modal, modal-table, strike and struck primitives: eleven measured bar/gong recipes with up to six…
Soft envelope gate: absolute-value detector, one-pole smoothed envelope with its own clamped time-coefficient policy, logistic comparator and VCA.
2-D and 3-D point, segment, plane and polygon routines ported from libroom geometry.cpp…
Glide in Modulation
Amplitude-dependent pitch glide as a per-quantum frequency multiplier: the Duffing-backbone initial detune, the reciprocal per-mode falloff, the one-pole envelope…
Large struck-plate gong voice (tam-tam, wind gong, daluo, xiaoluo, gong ageng, symphonic) as a modal resonator bank over the shared modal, modal-table, strike, struck, bloom…
Grain pool in Voices
Fixed-capacity grain pool (128 slots): start a grain at a source frame with a read step (negative reads backwards), a length, a grain-windows shape and a left/right gain pair…
The five grain windows of a granular voice (hann, tukey alpha 0.4, expodec 2% attack to -60 dB, reversed expodec, triangle) as a pure function of phase, a 512-point curve…
Allocation-free square/61-key hex key addressing shared by controller Processors (Isomorphic Keyboard, Exquis Controller).
Gillet Griesinger stereo reverb as CHOMPI ships it (Rings/Clouds topology): four input diffusers into a figure-eight loop inside one shared 32768-word ring, with exact…
H
I
J
Deterministic sample-and-hold reducer with sampling-clock jitter.
K
Canonical shared BS.1770 K filter coefficient and sample operations for offline/streaming consumers.
Key velocity response in Mixing
Register-interpolated gain and velocity exponent for reusable instrument response calibration.
Canonical keyed/sidechain dynamics compressor: widened soft-knee gain computer with a limiter sentinel, auto-makeup and stereo-link laws, bounded one-pole ballistics…
Key-map window predicates returning 0/1: whether a zone or layer admits (note, velocity byte) by its key and velocity ranges, whether its keyswitch window admits the selected…
L
Shared level primitives: floored f64 linear-amplitude-to-decibels conversion, f32 equal-power crossfade/pan gains, and an allocation-free note-transition audio-edge blend.
LFO in Modulation
Audio-rate low-frequency oscillator producing offset + depth * wave over the shared voice oscillator stages, so an LFO wave and an audio wave are the same curve at different rates.
Wet-only stereo sine LFO low-pass composed from one canonical LFO clock and independent canonical RBJ Biquad histories, using the documented Auto Filter exponential cutoff law.
Linear attack/sustain/release amplitude envelope over a bounded voice pool as functions over preallocated typed-array state: the f32 step law `fround(1 / (seconds * rate))`…
Stereo dual oscillator holding an exact integer frequency ratio from one f32 master accumulator, so a just-intonation Lissajous figure is a closed curve that cannot precess.
Mono and stereo linked true-peak lookahead brickwall limiter: 4x polyphase true-peak detector, sliding-minimum hold, two-pole release, boxcar attack, delayed apply and arithmetic…
Reading a bounded region of a host-loaded f32 sample table in play-order coordinates: reversal as a single read-address subtraction applied last…
Bounded paired stereo window matching, fresh learn/hold and continuous correction.
Bounded live summed-stereo K/RMS window meter independent of matching, signal gating, or gain control.
M
Beater contribution to a struck plate's spectrum: softness-driven per-mode spectral tilt, a bounded normalized half-sine contact-force pulse, and the constant-energy roughness…
Acoustic material records from the pyroomacoustics absorption (91) and scattering (8) tables: name-to-id lookup, flat-curve detection, octave-band resampling of coefficients…
Generated, read-only pyroomacoustics materials.json tables (absorption and scattering names, categories, offsets, lengths, coefficients and centre frequencies), pinned…
Prepared four-voice MIDI note maps with held-source snapshots and shared-output reference counting.
Bounded per-instance held-note step pattern shared by the Steps and Arpeggiator module Processors: steps-per-second rate converted to exact fractional frames through…
Bounded sample-clock MIDI timeline playback shared by prepared MIDI sources, including note release and loop transport.
Two-input mix operator: sum, linear dry/wet, equal-power crossfade and ring modulation with an equal-shape dry/wet.
Coupled-form (rotation) two-pole modal recurrence: f32 pole radius/angle, rotation coefficients, displacement and finite-difference velocity taps, per-mode amplitude roll-off…
Modal radiation in Physical models
Bounded modal body/readout stages: smooth high-pass-shelf radiation weight, displacement/velocity readout mix and symmetric per-channel gain, all clamped so control-plane model…
Modal resonator in Physical models
Single addressable two-pole modal resonator: T60-to-decay mapping, f32 rotation coefficient refresh, excite-rotate-emit tick and the gain/roll-off output stage.
Modal resonator bank in Physical models
Fixed-capacity (16-mode) modal resonator bank and packed multi-bank pool over preallocated typed-array state: per-mode pole configuration, bounded mode count, common-excitation…
Bounded struck-resonator voice shared by Bells, Handpan and Kalimba: a finite 128-slot note map with per-note bounded mode-weight vectors over an eight-recipe bank, a twelve-slot…
Modal table in Physical models
Publication-time modal-table preparation: recipe-major slot offsets, bounded mode counts, pitch-scaled mode frequency and decay, and the legacy readout roll-off gate.
Shared held-note selection and mono/legato retrigger decisions extracted from Vocoder and Phase 7.
Monophonic vocal-tract physical model: an LF glottal-flow source with aspiration, tenseness, throat tilt, vibrato and auto-wobble driving a 44-section Kelly-Lochbaum digital…
MIDI 1.0 MPE semantics (L2): zone configuration, per-channel bend range, member-channel note identity, and summed master/member pitch-bend, pressure and CC74 as CLAP-numbered…
Multi effect chain in Processors
Canonical reusable MultiFX effect-chain engine and numeric multi-effect-chain helpers, consumed by MultiFX and Loudmatch.
CHOMPI firmware port of the two-integrator resonant multimode filter (BasicMMF): band-pass feedback cascade with low/high/band-pass taps and a raw per-sample coefficient…
Musical clock: the shared conversion of beats (quarter notes), seconds, milliseconds and clock pulses to frames at the prepared sample rate, plus a tempo/beat-position clock…
N
NAM A2 in Processors
Bounded standard NAM A2 Lite/Full causal WaveNet inference using the original learned weight stream.
NAM A2 resource spec in Processors
Lightweight numerical prepared standard A2 resource layout shared by NAM pack writers and portable Processor readers.
Lightweight canonical standard NAM A2 topology, constants and weight-count contract for import validation.
Newlib rand in Noise
newlib rand(): the 64-bit LCG next = next * 6364136223846793005 + 1 returning bits 32..62, with the default seed 1 and srand semantics, as the CHOMPI TEMPO firmware draws it…
Note stack in Voices
Held-note stack for mono and legato play modes: held flags, per-note velocity, press order and last-note priority lookup.
O
Antoni octave-band weighting (pyroomacoustics AntoniOctaveFilterBank, pyfar window): band layout, index-exact bin edges, G_b(n) windows, exact G^2 spectral band…
Thresholded zero-crossing octave divider with envelope, tone and DC filtering.
One-pole low-pass tone-control primitives shared by the chain and the addressable tone unit: the bounded cutoff-to-coefficient mapping and the stateful per-sample update whose…
Just-intonation dual-oscillator stereo instrument voice over the shared stage primitives: a bounded eight-voice pool with same-note retrigger, lowest-free allocation…
Oversampled waveshaper in Distortion
Canonical fixed-4x oversampled waveshaper transfer, FIR state, tick and reset semantics shared by kernel nonlinear units.
DaisySP Adsr with exact f32 semantics: one-pole stages chasing overshooting targets with log time constants, cached setters, soft and hard retrigger, as the CHOMPI sample voices…
P
Equal-power stereo pan law over the shared crossfade gain pair, with a bounded position and plain double per-channel multiplies.
Parameter-macro table layout and value law: an f32 table of a version/row-count header row plus up to 32 [macroIndex, targetParamIndex, min, max, curveCode, reserved] rows that…
Eight-band-capable fused RBJ parametric EQ with bounded cascaded cut slopes, coefficient smoothing, transitions, tick and reset semantics.
Partitioned convolution in Filters
Uniformly partitioned overlap-save FFT convolver for long IRs ( packet 220-E): a frequency-domain delay line, a zero-latency direct head over filters/convolution…
Pattern arpeggiator in Processors
CHOMPI TEMPO firmware port of the two-engine ArpeggiatorSequencer: held and latched note lists with std::vector semantics, five step orders, five rest masks, random octave…
libDaisy f2s16/s162f signed 16-bit PCM conversions with exact f32 semantics: the +/-0.999985 clamp, truncation toward zero and the f32 1/32767 scale.
CHOMPI firmware port of the stereo int16-storage interpolated delay line: decrementing write pointer, linear reads truncated to int16 and the firmware's allocation/active-size…
CHOMPI firmware port of the smoothed PCM16 ping-pong delay with its dry/wet law: mono-summed input, cross-lane repeats, one-pole glided feedback/time/amount and the TAPE or WAVE…
Canonical feed-forward peak-compressor state, quadratic soft-knee curve, f32 gain-reduction ballistics, tick and reset semantics.
Pinned Peaks envelope function with fixed-point Twin ADSR/Split AD channel mapping, independent gates and explicit unsupported modes.
Fixed six-element permutation rank and unrank shared by serial DSP and host presentation.
Six-operator phase-modulation FM instrument voice with the complete Mark I parameter model: 32 fixed routings with multi-operator feedback closures, per-operator…
Phaser in Modulation
Four swept first-order allpass sections summed against dry, with a live relative phase-offset control for paired stereo lanes.
Speed of sound from temperature and humidity, temperature inversion, the octave-band air absorption table (7 bands, 6 humidity/temperature rows) and the shared DEFAULT_*…
Piano scale in Physical models
Provisional per-key piano length, core diameter, effective wound mass, tension, string count, hammer mass and impedance, with canonical tuning and stiff-string partial equations.
True ping-pong delay: two cross-feeding fixed lines producing structurally alternating L/R repeats, 100% wet.
Pitch-ratio math for sample players and tuned voices: exp2 through the natural exponential (`Math.exp(x * ln 2)`, the expression Maru compiler lowers directly), semitone and cent…
Streaming bounded YIN-lite monophonic pitch detector exposing frequency, period, confidence and voiced state per fixed hop.
Bounded monophonic TD-PSOLA pitch shifter over a fixed ring, with the input period as an explicit control input.
Causal monophonic note gating, continuous pitch and experimental circular tuning compensation.
Pinned Mutable Plaits Voice with original virtual-analog (firmware8) and additive (firmware12) engines.
Stereo low-band burst attenuation with rising-energy protection, manual reduction and processed/removed/dry audition.
Extended Karplus-Strong plucked-string voice (Jaffe & Smith): a raised-cosine pick pulse blended with seeded-xorshift32 one-pole-lowpassed noise, peak-normalized and combed…
PolyBLEP oscillator unit: sine/saw/square/triangle selection over the shared voice stage primitives, with octave/semitone/cent pitch offsets and an f32 phase accumulator.
Deterministic fixed-capacity voice ownership: same-note retrigger then lowest free slot then oldest active slot allocation, release marking, retirement and active/held counts.
Portamento in Voices
Exact-duration semitone glide extracted from Phase 7.
Explicit F64 numeric intent and the allocation-free identity helper asF64.
Primitives in Delays
Fixed-delay stage primitives shared by the chain and the addressable delay units: the millisecond-to-frames line length, the dry/wet output mix and the bounded feedback write value.
Continuous stereo gated integrated LUFS, momentary/short-term loudness, LRA, true peak and sample-timed dynamics/stereo history.
Q
R
Audio SDK-compatible f64 range policy: signed zero follows ECMAScript Math.min/Math.max, reversed bounds select the maximum, clamp propagates NaN and bounds infinities…
Rational soft clip in Distortion
DaisySP/stmlib rational SoftLimit x(27 + x^2)/(27 + 9x^2) and its +/-3-bounded SoftClip in the source's f32 operation order, as the CHOMPI firmware uses them.
Budgeted shoebox ray energy with absorption, scattering and receiver directivity, in fixed histogram buffers.
Closure-importable real FFT owner (/ decision): power-of-two real forward/inverse transforms up to 2^19 points in split re/im N/2+1-bin layout…
Deterministic crackle, dust and mains-hum generators with prepared scalar/filter state.
Reference smoothed gain in Processors
Qualification-only mono f32 smoothed-gain processor with deterministic reset.
Retune in Modulation
Pitch-correction control: retune speed, amount, humanize, transition dwell, unvoiced hold and delayed vibrato to a per-sample shift ratio.
Windowed reverse delay over caller-owned stereo rings with edge fades.
Ring stack in Math
Ring stacks of a heptatonic scale: line-of-thirds order, hinge, signed positions -3..+3, the diatonic stack at a position and height as pitch classes or close intervals, ring…
Rings modal-model resonator core with original parameter laws, bounded polyphony, separate odd/even taps and independent signal inputs.
Budgeted response display, Schroeder curve, broadband T30 and measured octave-band T30 over shared FIR and RT60 leaves.
Budgeted restartable shoebox RIR assembly over ISM, fractional-delay trains, octave weighting and zero-phase high-pass.
Image-source RIR builder (pyroomacoustics rir_builder.cpp): Hann-windowed sinc LUT, ranged f64 accumulate plus exact float32 emulation, delay_sum, fractional_delay…
Room Acoustics long-IR convolver (): two filters/partitioned-convolution two-stage zero-latency convolvers (128-tap head, 128-frame partitions, spread 2048/4096-frame tail)…
Deterministic round-robin take selection over a compile-time ceiling of 128 groups and 32 takes, with the rotation (cursor, visited mask, previous take) kept per group AND per…
numpy-compatible roundHalfEven (ties to the even integer, signed zero preserved) and C-style truncToward0 helpers used by the acoustics leaves.
Seeded Poisson late response weighted by ray energy histograms and shared octave FIRs.
RT60 theory (Sabine, Eyring) and inverse Sabine band absorption targets, plus Schroeder-integral RT60 measurement from an impulse response…
S
Sample rate reducer in Distortion
DaisySP SampleRateReducer (MIT, from Mutable Instruments Plaits): a phase accumulator re-samples the input at a 0..1 ratio of the host rate, places each held value at…
Bounded mono sample-table reader serving cyclic wavetable playback and one-shot regions with linear interpolation.
Sampler effect-bus table layout (): an f32 slot of a [version, busMask, zoneCount] header, three 105-value multi-effect-chain configurations for buses 2..4 and one bus code per…
Zone-mapped sampler reader over a host-loaded sample arena: selection by key range, velocity layer and per-zone/instrument keyswitch window, root-pitch and source-rate playback…
Canonical bounded tape-delay core: f32 damping pole, rational-tanh soft clip, bounded feedback write, dry/wet split and reset semantics.
Static scale masks, key/enable-mask/reference-A mapping and a hysteretic nearest-note quantizer with pass-through.
Canonical bounded four-comb/two-allpass Schroeder-Freeverb state, f32 damped feedback network, tick and full-tail reset semantics.
Serial effect stack in Processors
Bounded serial effect scheduling with stable slot identities, opaque type tags and queued live transitions.
Serial effect stack reference in Processors
Qualification-only restricted three-stage consumer of the serial effect scheduler.
Voice-slot selection by monotonic serial over a compile-time ceiling of 64 slots: the lowest free slot, else the active slot with the lowest age.
Bounded, resumable shoebox image-source model (pyroomacoustics image_source_shoebox): L1-ball walk up to order 17 (7175 images), per-band wall attenuations, per-mic directions…
CHOMPI TEMPO firmware port of the SimpleCompressor sidechain duck: a boosted, rectified key envelope above a 0.1 threshold lowers the gain of a second stereo signal, smoothed…
Slew in Modulation
Shared one-pole parameter smoothing and bounded scalar slew limiter for click-sensitive controls such as delay read positions.
Per-sample voice stage primitives: PolyBLEP residual and sine/saw/square/triangle oscillator samples, phase advance, one-pole envelope coefficient/step with shared stage…
Trapezoidal-integrator TPT state-variable filter with seven response modes, bounded host values and f32 coefficient/state parity.
Aligned four-stem prepared stereo clip playback and mixing.
Stereo auto panner in Modulation
Shared sine-LFO motion of a stereo input through the canonical constant-power stereo pan law.
Linked stereo peak/RMS compressor with independent signed-channel detector high-pass filters before the max-magnitude link.
Wet-only Dattorro adapter with bounded live predelay and normalized decay/damping controls.
Linked high-pass peak detector controlling a dynamic high shelf.
Linked stereo downward expander with a ratio-based quadratic soft knee, canonical finite peak detector and independent opening/closing ballistics.
Stereo high lowpass in Filters
Neutral wet-only stereo high-pass and low-pass sections composed from canonical RBJ Biquads, with independent lane histories and explicit zero-cutoff Off states.
Linked stereo peak compressor and hysteretic gate gain controllers composed from canonical detector and ballistics leaves.
Stereo ring modulator in Modulation
Wet-only stereo ring modulation over one phase-locked Lissajous carrier: the left lane uses the base frequency and the right lane may use a bounded integer ratio and static phase…
Wet-only stereo four-line FDN with physical decay time, live room-size slew and bounded staged clearing.
Wet-only dual-core Schroeder adapter with explicit right-channel decorrelation and bounded live predelay.
Stereo transient shaper in Dynamics
Stereo transient shaping from the pinned Gin full-band amplitude-envelope law, with independent detector and processing M/S balances.
Register-scaled one/two/three-string modal/shared-load/waveguide piano recipe with shared nonlinear impedance-load felt contact, bounded modal bandwidth and reduced secondary…
Per-mode excitation gains at the moment of contact: free-free bar and gong/kettle mode-shape weights with a mallet-width floor, plus the contact-time velocity tilt across the bank.
Prepared piano/string radiation readout with low cut, attenuating band and high shelf.
Instrument-independent struck-idiophone stages: squared velocity-to-gain curve, velocity-coupled effective hardness, parameterized contact-time burst length, exciter cutoff…
Aura grid eight-zone repeated chord-pitch selection.
Fused G1 subtractive voice over the shared stage primitives: schema-clamped oscillator/filter/envelope/gain parameters, full retrigger note-on and gate-owning note-off…
T
Four-tap, Haas and damped feedback-comb compositions over caller-owned stereo rings.
CHOMPI TAPE firmware looper: first-pass recording, varispeed -2..2 playback and overdub over one bounded interleaved int16-in-f32 buffer, dub-gain feedback through the hard…
Taper in Distortion
Tape processor: nine-control cook, companding encode/decode, flutter fractional delay, bias slew, mid/sub split, head-bump resonators and stereo soft clip.
Sixteen-pad morphing sample-playback voice: barycentric apex-base triangle morphing across up to three corner clips per pad with equal-power renormalization over loaded corners…
Web Audio `speakers` stereo-to-mono downmix: the arithmetic mean (left + right) * 0.5 at f32.
Addressable one-pole low-pass tone filter over the shared one-pole primitives, holding the derived coefficient and a double filter accumulator that is deliberately never frounded.
Tone-15.1 rational distortion transfer with an f32-rounded result, silence gate and a bounded drive amount.
Canonical non-amplifying sine tremolo state, f32 LFO phase accumulator, authored phase offset, tick and reset semantics.
Bounded static triode load-line solver and prepared tone/DC filtering.
Published BS.1770-5 four-phase true-peak detector and shared twelve-tap phase kernel.
U
V
DaisySP VariableShapeOscillator port: a hard-syncable saw/triangle/square morph with band-limited pulse-edge and wrap transitions bounded to two per sample.
Stride-3 Float64Array vector helpers (set, copy, add, sub, scale, dot, cross, norm, distance, normalize) replacing pyroomacoustics' Eigen Vector3f operations.
Velocity laws for sample players: the f32 squared playback gain, the MIDI byte a normalized velocity selects layers with (`floor(clamp(v * 127) + 0.5)`), and the equal-power…
Canonical fully wet sine-swept fractional-delay vibrato state, DelayNode-ordered tick and fresh-instance-identical reset semantics.
Clean-room channel vocoder: logarithmic band-pass analysis/synthesis bank over preallocated f32 state, per-band envelopes, internal carrier noise and an equal-power dry/wet split.
Voice pitch detector in Analysis
Voice YIN pitch detector with decimated analysis, full-rate refinement, voicing hysteresis and configurable 50–1600 Hz range and hop.
Monophonic TD-PSOLA voice pitch shifter for pitch correction: smoothed continuous ratio, independent formant factor, mark-spacing pitch, fixed declared latency…
Voiced overdrive in Distortion
Wet voiced drive composing canonical transfer, input low-cut, tone filtering and output trim.
Voiced wavefolder in Distortion
Wet wavefolder composed from the canonical Daisy fold, post-fold RBJ DC blocker, ToneFilter and output trim.
W
Planar 3-D polygon wall state: Newell normal and Gram-Schmidt in-plane basis (replacing upstream's SVD), area, segment intersection codes, reflection, side, same-as…
Warble in Modulation
CHOMPI firmware port of the Warble tape wow: a seeded LCG retargets a glided read time over two 1024-frame interpolated delay lines, blended with the dry signal by the frequency…
Pinned original Mutable Warps normal and frequency/phase-shifter core, including carrier crossmodulation and vocoder, in the qualified lookup domain.
DaisySP periodic triangular wavefold with signed pre-gain and offset.
Canonical mid/side stereo-width matrix state, bounded coefficient derivation and per-channel sample operations.
X
Xorshift32 in Noise
Raw unsigned 32-bit xorshift32 transition without seed normalization, plus the canonical unrounded mapping from a raw u32 to [-1, 1).
Z
Z in Oscillators
DaisySP ZOscillator port: a phase-distortion carrier contour multiplied by a formant discontinuity, with the discontinuity reset band-limited and shape/mode double-buffered.