k2k audio Volume control with a brain. Compressors, gates, and limiters react to how loud the signal is — taming peaks, opening up quiet parts, or clamping everything to a ceiling.
Native (Time-Domain)
These four work on the audio waveform directly — classic dynamics processing with character options and stereo linking.
Native Compressor
What it does — Reduces the volume of signals that exceed a threshold, with selectable character (Clean, VCA, Opto, FET) and soft-knee shaping.
When you’d reach for it — A vocal take has wide dynamic swings and you want it to sit more evenly in a mix, or a drum bus needs glue without touching the spectral balance.
Quick example
- Connect your audio source to the Compressor input.
- Set Threshold to around -20 dB so the loudest peaks trigger gain reduction.
- Dial Ratio to 4:1 for moderate leveling.
- Choose the FET character for an aggressive, punchy feel on drums (or Opto for smooth vocals).
- Adjust Dry/Wet to taste for parallel compression — keep some uncompressed signal blended in.
Parameters
| Parameter | What it controls | Range | Sweet spot hint |
|---|---|---|---|
| Threshold | Level where compression kicks in | -60 to 0 dB | -20 dB for general use |
| Ratio | How much gain reduction above threshold | 1:1 to 20:1 | 4:1 for moderate compression |
| Attack | How quickly compression engages | 0.1 to 100 ms | 10 ms balances punch and control |
| Release | How quickly compression lets go | 10 to 1000 ms | 100 ms for musical recovery |
| Knee | Transition softness around threshold | 0 to 24 dB | 6 dB for transparent onset |
| Makeup | Post-compression volume boost | 0 to 24 dB | Match perceived loudness to bypass |
| Sidechain HPF | Highpass on the detection signal | 20 to 500 Hz | 80 Hz to stop bass from pumping |
| Dry/Wet | Parallel compression blend | 0.00 to 1.00 | 0.50 for punchy parallel mixing |
| Detection | Peak or RMS level sensing | Peak / RMS | RMS for smoother, Peak for transient control |
| Character | Compression flavor | Clean / VCA / Opto / FET | FET for drums, Opto for vocals |
| Auto Makeup | Automatically compensate gain reduction | On / Off | On when exploring, Off for precise control |
| Link | Stereo linking mode | Dual / Link / Max | Link preserves stereo image |
Native Expander
What it does — Attenuates signals that fall below a threshold, widening the gap between loud and quiet parts.
When you’d reach for it — A recording has low-level bleed or room noise between phrases that you want pushed further down without gating it abruptly.
Quick example
- Feed the noisy recording into the Expander.
- Set Threshold to -30 dB, just above the noise floor.
- Use a Ratio of 2:1 for gentle expansion.
- Set Range to -40 dB to cap the maximum attenuation so it stays natural.
- Blend with Dry/Wet if you want partial noise reduction.
Parameters
| Parameter | What it controls | Range | Sweet spot hint |
|---|---|---|---|
| Threshold | Level below which expansion starts | -60 to 0 dB | -30 dB for room noise |
| Ratio | Expansion severity | 1:1 to 10:1 | 2:1 for natural reduction |
| Attack | How fast the expander opens | 0.1 to 100 ms | 5 ms to preserve transient starts |
| Release | How fast the expander closes | 10 to 1000 ms | 100 ms for smooth tail-offs |
| Range | Maximum attenuation depth | -80 to 0 dB | -40 dB for controlled reduction |
| Knee | Transition softness | 0 to 12 dB | Slight knee avoids harsh onset |
| Sidechain HPF | Highpass on detection signal | 20 to 500 Hz | 80 Hz to ignore low rumble |
| Dry/Wet | Parallel blend | 0.00 to 1.00 | 0.50 for subtle noise reduction |
| Detection | Peak or RMS level sensing | Peak / RMS | RMS for smoother behavior |
Native Gate
What it does — Silences the signal when it drops below a threshold, with a hold timer and hysteresis to prevent chattering.
When you’d reach for it — A snare mic is picking up hi-hat bleed and you want clean silence between hits, or a voiceover has audible breaths during pauses.
Quick example
- Send the drum mic to the Gate input.
- Set Threshold to -40 dB so only actual snare hits open the gate.
- Keep Attack at 1 ms so transients pass through immediately.
- Set Hold to 50 ms to catch the full body of each hit.
- Adjust Release to 100 ms for a natural fade-out after each hit.
Parameters
| Parameter | What it controls | Range | Sweet spot hint |
|---|---|---|---|
| Threshold | Level that opens the gate | -80 to 0 dB | -40 dB for drum close mics |
| Attack | How fast the gate opens | 0.1 to 50 ms | 1 ms to keep transients intact |
| Hold | Time the gate stays open after signal drops | 0 to 500 ms | 50 ms for natural sustain |
| Release | How fast the gate closes after hold | 10 to 1000 ms | 100 ms for smooth fadeout |
| Range | Attenuation when gate is closed | -80 to 0 dB | -80 dB for full silence |
| Hysteresis | Buffer zone below threshold to prevent flutter | 0 to 10 dB | 3 dB stops chattering |
| Sidechain HPF | Highpass on detection signal | 20 to 500 Hz | 80 Hz to ignore kick bleed |
| Dry/Wet | Parallel blend | 0.00 to 1.00 | 0.50 for subtle gating |
| Detection | Peak or RMS level sensing | Peak / RMS | Peak for drum transients |
Native Limiter
What it does — Brickwall peak limiter that prevents any signal from exceeding the ceiling, with optional lookahead for transparent transient handling.
When you’d reach for it — You are finalizing a mix and need to guarantee nothing clips, or you want to push loudness without distortion.
Quick example
- Place the Limiter at the end of your signal chain.
- Set Threshold to -1 dB as your absolute ceiling.
- Leave Release at 100 ms for transparent recovery.
- Enable Lookahead at 2-3 ms if transients sound crunchy.
- Check the GR meter — more than 3-4 dB of constant reduction means you are pushing too hard.
Parameters
| Parameter | What it controls | Range | Sweet spot hint |
|---|---|---|---|
| Threshold | Maximum output level (ceiling) | -24 to 0 dB | -1 dB for safe headroom |
| Release | Recovery time after limiting | 10 to 500 ms | 100 ms for general use |
| Lookahead | Anticipate peaks for cleaner limiting | 0 to 5 ms | 2 ms for transparent transients |
| Dry/Wet | Parallel blend | 0.00 to 1.00 | 0.50 for parallel limiting |
| Link | Stereo linking mode | Dual / Link | Link to preserve stereo image |
Spectral (Per-Frequency-Band)
These work in the frequency domain — the spectrum is split into bands and each band gets its own independent dynamics processing. You can also draw frequency curves to set different thresholds, attack, or release per band.
Spectral Compressor
What it does — Compresses each frequency band independently, so you can tame harsh highs without touching warm lows.
When you’d reach for it — A vocal is sibilant but also needs body compression, or a full mix has a few frequency ranges poking out inconsistently across the song.
Quick example
- Connect audio into the Spectral Compressor.
- Set Threshold to -20 dB and Ratio to 4:1.
- Increase Bands to 16 for finer frequency resolution.
- Enable Frequency Curves and draw the threshold lower in the 4-8 kHz range to de-ess.
- Use the Metering section to see which bands are working hardest.
Parameters
| Parameter | What it controls | Range | Sweet spot hint |
|---|---|---|---|
| Threshold | Level where compression starts per band | -92 to 0 dB | -20 dB for general compression |
| Ratio | Compression amount per band | 1:1 to 20:1 | 4:1 for moderate control |
| Attack | How fast compression engages per band | 0.1 to 100 ms | 10 ms for musical response |
| Release | How fast compression releases per band | 10 to 1000 ms | 100 ms for smooth recovery |
| Makeup | Post-compression gain boost | 0 to 24 dB | Match perceived loudness |
| Knee | Transition softness | 0 to 12 dB | 6 dB for transparent onset |
| Bands | Number of frequency bands | 1 to 32 | 8 for general, 16+ for surgical |
| Detection | Peak or RMS level sensing | Peak / RMS | RMS for smoother results |
| Frequency Curves | Per-band threshold, attack, release curves | Toggle + draw | Draw threshold lower at sibilant frequencies |
Spectral Expander
What it does — Attenuates quiet signals independently per frequency band, pushing noise below threshold further down in each part of the spectrum.
When you’d reach for it — A recording has broadband noise but the noise floor varies with frequency — hiss up high, hum down low — and a single expander threshold cannot address both.
Quick example
- Feed the noisy recording into the Spectral Expander.
- Set Threshold to around -40 dB, just above the noise floor.
- Set Ratio to 2:1 for gentle expansion.
- Enable Frequency Curves and use the Noise Gate preset to target hiss in the highs.
- Adjust Range to -40 dB so expansion does not over-attenuate wanted content.
Parameters
| Parameter | What it controls | Range | Sweet spot hint |
|---|---|---|---|
| Threshold | Level below which expansion starts per band | -92 to 0 dB | Just above the noise floor |
| Ratio | Expansion severity per band | 1:1 to 20:1 | 2:1 for natural reduction |
| Attack | How fast the expander opens per band | 0.1 to 100 ms | 5 ms for responsive tracking |
| Release | How fast the expander closes per band | 10 to 1000 ms | 50 ms for smooth fades |
| Range | Maximum attenuation depth | -96 to 0 dB | -40 dB for controlled reduction |
| Knee | Transition softness | 0 to 12 dB | 6 dB for transparent behavior |
| Bands | Number of frequency bands | 1 to 32 | 8 for general, 16+ for surgical |
| Detection | Peak or RMS level sensing | Peak / RMS | Peak for faster response |
| Frequency Curves | Per-band threshold, attack, release curves | Toggle + draw | Use presets: Noise Gate, Transient, De-Reverb |
Spectral Gate
What it does — Gates each frequency band independently, silencing quiet bins while letting loud ones pass — like having a separate noise gate on every slice of the spectrum.
When you’d reach for it — You want to gate hiss in the highs and rumble in the lows with different thresholds, or clean up a recording where noise sits at specific frequencies.
Quick example
- Connect the noisy audio to the Spectral Gate.
- Set Threshold to -50 dB and Range to -80 dB.
- Set Hysteresis to 3 dB to prevent gate flutter.
- Enable Frequency Curves and load the Hiss Gate preset for high-frequency cleanup.
- Increase Bands to 16 for finer control over which frequencies gate.
Parameters
| Parameter | What it controls | Range | Sweet spot hint |
|---|---|---|---|
| Threshold | Level that opens the gate per band | -92 to 0 dB | -50 dB for general noise |
| Attack | How fast the gate opens per band | 0.1 to 50 ms | 1 ms to preserve transients |
| Hold | Time the gate stays open after signal drops | 0 to 500 ms | 50 ms for natural sustain |
| Release | How fast the gate closes per band | 10 to 500 ms | 50 ms for smooth closure |
| Range | Attenuation when gate is closed | -96 to 0 dB | -80 dB for near-silence |
| Hysteresis | Buffer zone to prevent gate flutter | 0 to 12 dB | 3 dB stops chattering |
| Bands | Number of frequency bands | 1 to 32 | 8 for general, 16+ for surgical |
| Detection | Peak or RMS level sensing | Peak / RMS | Peak for transient accuracy |
| Frequency Curves | Per-band threshold, attack, release curves | Toggle + draw | Use presets: Hiss Gate, Rumble Gate, Drum Gate |
Spectral Gating
What it does — Uses one signal (Gate) to control which frequency bins of another signal (Main) pass through, on a per-bin basis.
When you’d reach for it — You want a vocal to only come through where a synth is active, creating a vocoder-like effect, or you want to extract only the frequencies where two recordings overlap.
Quick example
- Connect the synth pad to the Gate input and the vocal to the Main input.
- Set Threshold to -40 dB so only prominent frequencies in the synth open the gate.
- Adjust Attack to 5 ms and Release to 50 ms for smooth transitions.
- Try Inverse mode to hear the vocal only where the synth is quiet.
- Use Smoothing to reduce bin-to-bin artifacts in the gated output.
Parameters
| Parameter | What it controls | Range | Sweet spot hint |
|---|---|---|---|
| Threshold | Gate signal level that opens bins for Main | -92 to 0 dB | -40 dB for prominent features |
| Range | Attenuation applied to closed bins | -96 to 0 dB | -60 dB for strong gating effect |
| Attack | How fast bins open | 0.1 to 100 ms | 5 ms for responsive tracking |
| Release | How fast bins close | 10 to 500 ms | 50 ms for smooth tails |
| Mix | Blend between dry Main and gated result | 0 to 100% | 100% for full effect |
| Output Gain | Post-gating volume adjustment | -24 to 24 dB | 0 dB unless compensating |
| Mode | Normal (Gate opens Main) or Inverse (Gate closes Main) | Normal / Inverse | Normal for masking, Inverse for ducking |
| Detection | Peak or RMS level sensing | Peak / RMS | Peak for sharp gating |
| Lookahead | Frames of anticipation | 0 to 8 | 0 for zero latency, 2-3 for cleaner edges |
| Smoothing | Frequency-axis smoothing of gate gains | 0.00 to 1.00 | Light smoothing reduces zipper artifacts |
Spectral Ducker
What it does — Ducks the main signal per frequency band whenever the sidechain signal gets loud, letting one sound carve space for another at specific frequencies.
When you’d reach for it — A pad is masking the vocal in the midrange and you want the pad to duck only in that frequency region when the voice comes in, while the pad’s low end stays untouched.
Quick example
- Connect the pad to the Main input and the vocal to the Sidechain input.
- Set Threshold to -20 dB so the vocal triggers ducking.
- Set Range to -12 dB for noticeable but not extreme ducking.
- Enable Frequency Curves and raise the threshold in the lows so only the midrange ducks.
- Use Hold at 50 ms to keep the duck smooth between syllables.
Parameters
| Parameter | What it controls | Range | Sweet spot hint |
|---|---|---|---|
| Threshold | Sidechain level that triggers ducking | -92 to 0 dB | -20 dB for vocals |
| Range | How much the main signal drops when ducked | -96 to 0 dB | -12 dB for musical ducking |
| Attack | How fast the duck engages | 0.1 to 100 ms | 5 ms for responsive ducking |
| Hold | Time the duck stays active after sidechain drops | 0 to 500 ms | 50 ms between syllables |
| Release | How fast the signal recovers | 10 to 1000 ms | 100 ms for smooth return |
| Bands | Number of frequency bands | 1 to 32 | 8 for general, 16+ for surgical |
| Detection | Peak or RMS level sensing | Peak / RMS | RMS for smoother response |
| Sidechain Listen | Preview what the sidechain hears | On / Off | On while setting threshold |
| Frequency Curves | Per-band threshold, attack, release curves | Toggle + draw | Raise threshold where you want no ducking |
Spectral Limiter
What it does — Brickwall limiter applied independently per frequency band, so nothing in any band exceeds the ceiling.
When you’d reach for it — A mix has occasional harsh peaks in the upper mids that clip before the rest of the spectrum is loud enough, or you want frequency-aware loudness maximization.
Quick example
- Connect your mix to the Spectral Limiter.
- Set Ceiling to -1 dB as your target maximum.
- Push Input Gain up by a few dB to drive more signal into the limiter.
- Enable Frequency Curves and use the Protect Highs preset to limit the high end more aggressively.
- Watch the Metering section to verify even gain reduction across bands.
Parameters
| Parameter | What it controls | Range | Sweet spot hint |
|---|---|---|---|
| Ceiling | Maximum output level per band | -24 to 0 dB | -1 dB for safe headroom |
| Release | Recovery time after limiting per band | 10 to 1000 ms | 100 ms for transparent recovery |
| Input Gain | Pre-limiter drive | -12 to 24 dB | 0 dB, increase to push loudness |
| Bands | Number of frequency bands | 1 to 32 | 8 for general, 16+ for surgical |
| Frequency Curves | Per-band ceiling and release curves | Toggle + draw | Use presets: Protect Highs, Bass Control, Loudness Max |
Spectral Transient Shaper
What it does — Boosts or cuts the attack and sustain portions of the signal independently per frequency band, reshaping the dynamics of each part of the spectrum.
When you’d reach for it — A drum loop needs more snap in the upper mids but the low-end body should stay pillowy, or you want to tighten bass transients without affecting the rest.
Quick example
- Feed a drum loop into the Spectral Transient Shaper.
- Boost Attack to +6 dB for more snap.
- Cut Sustain to -3 dB to tighten the tail.
- Enable Frequency Curves and load the Punchy Drums preset for frequency-shaped transient emphasis.
- Adjust Sensitivity to control how easily the detector flags a transient.
Parameters
| Parameter | What it controls | Range | Sweet spot hint |
|---|---|---|---|
| Attack | Gain applied to transient portion per band | -24 to +24 dB | +6 dB for punch, -6 dB for softness |
| Sustain | Gain applied to sustain portion per band | -24 to +24 dB | -3 dB to tighten, +3 dB to swell |
| Sensitivity | How easily a transient is detected | 0 to 100% | 50% for general use |
| Output Gain | Post-shaping volume adjustment | -12 to +12 dB | 0 dB unless compensating |
| Fast | Fast envelope tracking time | 0.1 to 10 ms | 2 ms for percussive content |
| Slow | Slow envelope tracking time | 10 to 200 ms | 80 ms for stable sustain reference |
| Bands | Number of frequency bands | 1 to 32 | 8 for general, 16+ for surgical |
| Frequency Curves | Per-band attack and sustain curves | Toggle + draw | Use presets: Punchy Drums, Soft Attack, Tight Bass |