KERN SMOOTH vs ANINA: two approaches to resonance suppression

why this comparison matters

a harsh vocal lands in your session. you need a resonance suppressor, but you are not sure which one. ANINA from CRQL is free. KERN SMOOTH is $29. soothe 2 is $219.

the price gap is obvious. the engineering difference is less so. and the engineering is what actually matters for your mix.

ANINA and KERN SMOOTH solve the same problem with fundamentally different approaches. ANINA uses up to 1024 linear frequency bands. KERN SMOOTH uses 40 ERB bands mapped to human hearing. more bands sounds better on paper. but your ears do not hear in linear frequency, and that distinction changes everything about how these two tools behave in practice.

full disclosure: i make KERN SMOOTH. this comparison gives ANINA the same honest treatment i give my own plugin. if ANINA is better for your use case, i will say so.

two approaches to the same problem

both ANINA and KERN SMOOTH analyze your audio in the frequency domain and apply dynamic gain reduction to resonant peaks. the difference is how they divide the frequency spectrum.

ANINA: up to 1024 linear bands

ANINA splits the spectrum into equally-spaced frequency bands using a standard FFT. a dropdown in the header lets you choose the block size: 2048 (the default, giving 1024 bands) down to 512 (256 bands, faster response). each band covers the same width in Hz regardless of where it sits in the spectrum.

this is straightforward and effective. every part of the spectrum gets the same resolution. if there is a resonance at 500 Hz or at 5 kHz, ANINA can find it with the same precision. the tradeoff is latency vs resolution: larger block sizes give finer frequency detail but add more latency.

KERN SMOOTH: 40 ERB bands

KERN SMOOTH uses 40 bands spaced according to the ERB (equivalent rectangular bandwidth) scale.[^1] ERB models how the human auditory system actually resolves frequency. at 1 kHz, one ERB is about 130 Hz wide. at 8 kHz, it is about 960 Hz wide.

the result: KERN SMOOTH has roughly 7x more resolution in the 1-5 kHz range (where your ears are most sensitive and where harshness lives) and less resolution in the extreme highs (where your ears cannot distinguish fine detail anyway).

what ANINA does well

ANINA is genuinely impressive for a free plugin. CRQL released it in early 2026 and it immediately became one of the most talked-about free tools in the resonance suppression category.

more controls than you expect for free. Amount sets suppression depth, Attack and Release control how fast the spectral compressor responds, and an expandable panel reveals S/C (sidechain input), LO/HI cutoff, and a Gate knob. a Freeze button locks the spectral analysis, and Delta lets you hear only the removed signal. the block size dropdown (512 to 2048) lets you trade off between temporal response and frequency resolution.

CLAP support. ANINA is one of the few resonance suppressors available in the CLAP format alongside VST3 and AU. if you use Bitwig or another CLAP-native host, this matters.

up to 1024 band resolution. at the default 2048 block size, ANINA operates with 1024 linear frequency bands. drop to 512 for faster response (256 bands) at the cost of frequency resolution. the flexibility is welcome.

sidechain and delta. external sidechain input lets you use one source to drive suppression on another. delta monitoring lets you hear exactly what is being removed. both are features that KERN SMOOTH does not currently offer.

zero cost, no strings. no iLok, no account, no subscription. download and use.

where it falls short. without psychoacoustic weighting, ANINA applies the same suppression sensitivity across the entire spectrum. a narrow resonance at 500 Hz (where your ears can distinguish ~65 Hz differences) gets the same treatment as one at 10 kHz (where your ears can only distinguish ~1000 Hz differences). this can mean over-processing in the highs and under-processing in the critical midrange, especially on complex material like vocals or full mixes. and while the control set is generous, there is no M/S routing and no dedicated transient mode.

what KERN SMOOTH does differently

KERN SMOOTH is built on a 5-stage spectral pipeline that processes audio through the ERB perceptual model at every step.[^1]

psychoacoustic frequency weighting. the ERB filterbank means SMOOTH is inherently more aggressive in the 1-5 kHz speech/harshness zone and more conservative outside it. you do not need to set this up. it is baked into the architecture.

dual processing modes. resonance mode targets sustained tonal resonances (room modes, feedback, ringing strings). transient mode targets percussive harshness (cymbal hits, plosives, digital clipping transients). ANINA uses a single mode.

M/S routing. process mid and side channels independently. harshness in the center of a mix (vocals, snare) often needs different treatment than harshness in the sides (cymbals, room ambience). ANINA processes stereo as a unit.

spectral reassignment. estimates the true instantaneous frequency of each component for more precise targeting than standard FFT bin resolution alone.[^3]

where it falls short. $29 is not free. no CLAP format (yet). no AAX for Pro Tools. smaller preset library than established tools. newer product with less track record.

the comparison

when to choose which

choose ANINA when:

• your budget is zero and you need capable resonance suppression now
• you need sidechain input or delta monitoring (KERN SMOOTH does not have these yet)
• you use a CLAP-native host like Bitwig
• you want adjustable block size for trading off latency vs resolution

choose KERN SMOOTH when:

• you work with vocals, acoustic instruments, or complex mixes where the 1-5 kHz range is critical
• you need separate resonance and transient modes
• you use M/S processing (treating center and sides differently)
• you want psychoacoustic frequency weighting without manual setup
• $29 is within your budget

choose soothe 2 when:

• you need AAX for Pro Tools
• you want the deepest feature set and largest preset library
• you need sidechain input with full routing flexibility
• $219 is within your budget

listening scenarios

harsh vocals

vocals sit squarely in the 2-5 kHz sensitivity zone. sibilance, nasal resonances, and proximity effect all concentrate energy where your ears are most critical. this is where ERB weighting makes the biggest difference: KERN SMOOTH naturally applies more sensitivity to these frequencies without you touching a control. ANINA will handle it too, but you may find yourself reaching for more suppression to get the same perceived result.

ringing drums

a snare with a 400 Hz ring or a kick with a boxy 200 Hz buildup. both tools handle this well. the resonance is narrow and easy to detect regardless of frequency scale. for isolated drum tracks, ANINA’s free price is hard to beat.

mix bus and mastering

full mixes contain energy across the entire spectrum. subtle over-processing in the highs (from equal-weight linear bands) can dull the air and presence of a mix. ERB weighting preserves the high-frequency character by applying less sensitivity where your ears cannot resolve fine detail anyway. for mix bus work, KERN SMOOTH’s resonance mode is designed for this exact use case.

frequently asked questions

frequently asked questions

is ANINA a good free alternative to soothe 2?

ANINA is a capable free resonance suppressor with up to 1024 linear bands, sidechain input, delta monitoring, and CLAP support. it handles resonance problems well, especially on straightforward material. for complex sources where psychoacoustic precision matters, tools like KERN SMOOTH ($29) offer ERB-weighted processing, dedicated transient mode, and M/S routing.

what is the difference between ERB bands and linear bands?

linear bands divide the frequency spectrum into equal-width segments. ERB (equivalent rectangular bandwidth) bands model human hearing, with narrower bands where your ears are most sensitive (1-5 kHz) and wider bands where they are not. this means ERB-based processing naturally focuses on the frequencies that matter most to perception.

is KERN SMOOTH worth $29 when ANINA is free?

it depends on your needs. ANINA is excellent for basic resonance suppression. KERN SMOOTH adds psychoacoustic ERB processing, dedicated transient mode, M/S routing, and a real-time spectral display. if you work with vocals, acoustic instruments, or mix bus processing where subtle perception matters, the ERB approach can make an audible difference.

does ANINA support VST3 and AU formats?

ANINA supports VST3, AU, and CLAP formats on macOS, Windows, and Linux. KERN SMOOTH supports VST3 and AU on macOS and Windows. neither currently offers AAX for Pro Tools.

which resonance suppressor uses less CPU?

both ANINA and KERN SMOOTH are designed for low CPU usage. KERN SMOOTH targets under 3% at 44.1 kHz with a single stereo instance. ANINA is similarly efficient. for most sessions, neither will be a bottleneck.

references

a note from the developer

this comparison exists because ANINA is a good plugin that deserves honest evaluation, not dismissal. CRQL built something useful and gave it away for free. that is worth respecting.

the reason KERN SMOOTH costs $29 is the ERB architecture underneath. mapping resonance detection to human auditory filters is not a marketing feature. it is four years of reading psychoacoustics papers, building prototypes that sounded wrong, and learning why equal-loudness contours matter for signal processing, not just mixing decisions.

if ANINA solves your problem, use it. if you find yourself pushing it harder than you would like and the highs are getting dull while the mids still ring, that is the gap psychoacoustic processing fills.

i am a solo developer in copenhagen. if you want to discuss any of this, reach out at jonas@kernaudio.io.