What it is
A resonance suppressor is a dynamic spectral processor: it continuously analyzes the spectrum, identifies the zones that stick out relative to their neighbors (resonances, harshness, sibilance, room peaks) and attenuates them only at the moment they appear, with hundreds of narrow “cuts” that move over time. It is an evolution of the dynamic EQ: instead of one or two bands set by hand, the processor decides on its own where to intervene within the limits you set.
The best known is oeksound soothe2; alternatives: Waves Curves Resolve, Sonible smart:EQ/smart:deess, iZotope Neutron Unmask, TDR Nova (with multiple dynamic bands), Baby Audio Smooth Operator, Tokyo Dawn Labs Prism.
Parameters (soothe2)
- Depth: amount of reduction. It is the main control: start low and raise it until you hear the harshness disappear.
- Sharpness: how narrow and deep the cuts are. High for pinpoint resonances (whistles, ringing), low for broad harshness.
- Selectivity: selection threshold. High = it intervenes only on the most obvious resonances; low = it intervenes on many zones and “smooths” the spectrum.
- Attack and Release: reaction speed. Fast attack for sibilance and harshness; longer release for a less grainy result.
- Soft/hard mode: soft is less level-dependent and more transparent; hard depends more on level and can be set more precisely.
- Band shaping: low cut, high cut and four bands that determine where the processor is more or less sensitive. Essential: it limits the action to the zone of the problem.
- Mid/Side, stereo link, balance: L/R or M/S processing and degree of linking between the channels.
- Delta: listen only to what is being removed. If in the delta you hear the whole vocal and not just the harshness, you are removing too much.
- Mix and trim: blend with the clean signal and compensation for comparison.
- Sidechain: the reduction is calculated from another track (e.g. the vocal controls the instrumental bus).
- Oversample and resolution: calculation precision, useful on the bass and with high sharpness.
Starting settings
| Source | Zone (band shaping) | Depth | Notes |
|---|---|---|---|
| Lead vocal | 2–8 kHz | low-medium | Harshness and sibilance; fast attack |
| Vocal recorded in a small room | 200–800 Hz | medium | Room resonances; longer release |
| Distorted guitars | 2–6 kHz | medium | Fizz and harshness of the wall |
| Acoustic guitar | 150–400 Hz and 2–5 kHz | low | Soundbox boom, harsh strings |
| Cymbals and overheads | 3–10 kHz | low | Harshness |
| Synths and supersaw | 1–6 kHz | medium | Filter and unison resonances |
| Instrumental bus (sidechain from the vocal) | 1–5 kHz | low | Dynamic space for the vocal |
| Mastering | 2–6 kHz | very low | Only obvious harshness in the mix |
Where in the chain
- After the static corrective EQ (HP and fixed cuts) and before compression, if the resonances make the compressor react.
- After compression, if it is the compression that brings out harshness (frequent on heavily compressed vocals).
- On distorted guitars and synths, after the distortion.
- In mastering, at the start of the chain and with minimal amounts.
Common mistakes
- High depth across the whole band: the sound becomes dull, “muffled”, without presence. The delta shows that the tone is being removed too.
- Using it as a substitute for arrangement: if three instruments occupy the same zone, the suppressor dulls them all.
- Selectivity too low: the spectrum is flattened and the sound loses character.
- No band shaping: the processor intervenes even where it is not needed (e.g. on the bass of the vocal).
- Judging without trim: the reduction lowers loudness, so comparing without compensation is skewed.
Starting settings
| Parameter | Value | Why |
|---|---|---|
| Depth | The minimum that eliminates the harshness | Beyond that, the sound becomes dull. |
| Selectivity | Medium-high | Intervenes only on obvious resonances. |
| Band shaping | Limited to the zone of the problem | No unnecessary interventions. |
| Mode | Soft to start with | More transparent. |
| Delta | Periodic listening | Check on what is being removed. |
| Mastering | Very low depth, 2–6 kHz | Only harshness from the mix. |
Frequencies to watch
| Range | What to do |
|---|---|
| 200–800 Hz | Room resonances, acoustic guitar boom: long release, medium depth. |
| 2–5 kHz | Harshness of vocals, guitars, synths: main zone of intervention. |
| 5–9 kHz | Sibilance and cymbal harshness: fast attack. |
| Above 10 kHz | Minimal intervention: removing here takes away air. |
Example chain
| Step | Tool | Settings | Why |
|---|---|---|---|
| Corrective EQ | Fruity Parametric EQ 2 | HP and static cuts | The suppressor works only on what varies. |
| Suppressor | soothe2 / Smooth Operator / Curves Resolve | Low-medium depth, band shaping 2–8 kHz | Dynamic resonances. |
| Compression | Track compressor | 3–6 dB | Level without exaggerating harshness. |
| De-esser | Dedicated de-esser or a second instance of the suppressor | 5–9 kHz | Residual sibilance. |
| Color | EQ and saturation | Presence and air | Final tone. |
How to check the result
- Track spectrum: narrow peaks more than 6 dB above neighboring bands reduced below 3 dB.
- High-mid band (2–4 kHz): relative energy reduced by 0.5–2 dB, no more (beyond that the sound becomes dull).
- Air band (8–16 kHz) unchanged within 1 dB: the suppressor must not take away the air.
- Track loudness compensated with the trim before comparison (difference under 0.5 LU).