Vocal Chain
Burnish Vocal Chain
The vocal chain that takes the record as an input.
A lead-vocal chain in one window: tune, clean, shape, compress, saturate, double, place and ride, each of the eight stages with its own scope. Drop a reference record and the chain proposes settings to sit your voice the way that record sits, showing its reasoning per stage with honest confidence labels. Analysis runs fully local.
Signal path
Why it exists
Every stage of a vocal chain is a guess about how the record should sit: how much tuning, where the harshness lives, how hard the compressor leans, how much air, how much room. Burnish puts all eight stages on one plate, gives each its own scope so you can see what it is doing, and lets a reference record make the first pass. Drop an acapella or a full mix and a fully local analysis extracts its long-term tonal curve, dynamics density, sibilance profile, space estimate and tune style, fits interpretable rules, and explains each move in plain language with a confidence label. Match Amount scales the whole fit from 0 to 150 percent. You accept, adjust or discard. It proposes settings so your voice sits the way the record sits. It never promises an identical sound.
Eight stations, one plate
Every stage has a scope. The record sets the first pass.
01 · Play
A real take, drawn as light.
The Play view puts the performance on the stage spectrum with its pitch trace, the gain reduction and the short-term loudness beside it. The header reads the active mode and the exact latency the plugin reports for it, at all times.
02 · Track
Track mode: 0.1 ms.
Track mode runs zero-lookahead dynamics on the Classic tune path with 4 samples of total latency at 48 kHz, test-enforced under a 5 ms contract, so the singer hears the chain while recording. Mix mode is full quality and reports its latency exactly, 12.5 ms with the Natural path.
03 · Reference
Drop the record. Read the reasoning.
An acapella arrives with full confidence; a full mix is isolated and labeled reduced confidence. The analysis extracts a 40-band long-term tonal curve, dynamics density, a sibilance profile, a space estimate with its own confidence label and a three-position tune style, then fits interpretable rules and shows what it did and why. Measured: 18 of 18 eval cells closer than flat on 6 real mixes and 3 real takes. No cloud, no account, ever.
04 · Tune
Two tuning paths, in-house.
Classic is the hard-tune sound at 0.1 ms. Natural is formant-preserving correction with a Throat control, holding the third formant within about 4 percent at a forced 3 semitone shift. Sibilants are never shifted, and the tracker made zero octave errors across the test batteries, with unvoiced detection at 100 percent on the noise suite.
05 · Scope
FET into Opto, on its own scope.
The COMP station: FET into Opto compression with each stage's gain reduction drawn on the station's own scope. Every lit element on the screen maps to a measured signal quantity, so what you see is the chain, not a picture of one.
Where it earns its keep
Built for the jobs you actually have.
Tracking a lead vocal
Track mode at 0.1 ms with the Classic path: the singer hears the tuned, compressed, placed chain in the headphones with no perceptible delay.
Mixing to a reference
Drop the record the artist brought in, read the per-stage reasoning, set Match Amount, and finish by ear with every stage still visible.
A chain that adapts per singer
Every one of the 66 factory chains carries a measured target profile; loading one arms Adapt to my voice, so the same chain sits a different voice differently.
Learning what each stage does
12 Learn chains isolate one stage each, with its scope lit, so the de-harsh or the rider can be heard on its own before it joins the chain.
What's in the unit
Real pixels, not promises.
The stage spectrum
The performance drawn as light over the stage spectrum, with the pitch trace, gain reduction and short-term loudness beside it.
Track mode, 0.1 ms
Zero-lookahead dynamics on the Classic path, 4 samples at 48 kHz, test-enforced under a 5 ms contract. The header reports it exactly.
Reference analysis, fully local
Tonal curve, dynamics density, sibilance profile, space estimate and tune style, fitted into interpretable settings with confidence labels.
The TUNE station
Classic or Natural path, Throat, key and scale, and the note keyboard, with the station's own scope showing the correction.
A scope on every station
Each of the eight stations carries its own scope. Every lit element maps to a measured signal quantity from the chain.
Eight stages, from the line's shipped DSP
| Stage | What runs there | Lineage |
|---|---|---|
| Tune | Classic hard-tune at 0.1 ms, or Natural formant-preserving correction with Throat | In-house |
| Clean | Per-band adaptive de-harsh that cuts resonant excess, not brightness; program-adaptive split-band de-ess | In-house |
| Shape | Five-band equalizer | Bevel EQ math |
| Compress | FET into Opto | Caliper Compressor |
| Color | Tape, tube and present color, plus air | Caliper and Iron Summing |
| Double | Detune-decorrelated doubler, mono fold-down under 3 dB | In-house |
| Space | Delay and reverb, both ducked from the vocal | Colonnade Delay and Halo Reverb |
| Ride and limit | Feedforward BS.1770 short-term loudness rider, plus or minus 0.5 LU convergence, then a safety limiter | In-house |
Everything in the box
- Eight stages in one window: tune, clean, shape, compress, color, double, space, ride and limit
- Each station carries its own scope; every lit element maps to a measured signal quantity
- Track mode: zero-lookahead dynamics, Classic path, 0.1 ms total latency (4 samples at 48 kHz)
- Mix mode: full quality, latency reported exactly in the header (12.5 ms with the Natural path)
- Classic tune path: the hard-tune sound at 0.1 ms
- Natural tune path: formant-preserving correction with Throat; sibilants never shifted
- Tracker measured at zero octave errors across the test batteries
- Reference analysis: acapella or full mix, offline, background, fully local
- Extracted: 40-band K-weighted tonal curve, dynamics density, sibilance profile, space estimate, tune style
- Interpretable fits with plain-language reasoning and confidence labels per stage
- Match Amount scales the whole fit from 0 to 150 percent
- Per-band adaptive de-harsh and program-adaptive split-band de-ess
- FET into Opto compression; tape, tube and present color with air
- Detune-decorrelated doubler with mono fold-down under 3 dB
- Delay and reverb, both ducked from the vocal
- BS.1770 short-term loudness rider and safety limiter
- 66 factory chains with measured target profiles; Adapt to my voice
- 12 Learn chains that isolate one stage each
- 131 parameters, add-only; state versioned from day one
- Mono or stereo in, stereo out; a lead-vocal chain by design
- Passes pluginval at strictness level 5 in demo and licensed states
- VST3, AU and AAX on Windows 10+ and macOS, Intel and Apple Silicon
Specifications
Questions engineers ask
Does the reference make my vocal sound identical to the record?
No. The analysis proposes interpretable settings so your voice sits the way the record sits, and shows its reasoning per stage with confidence labels. You accept, adjust or discard, and Match Amount scales the whole fit from 0 to 150 percent.
Does reference analysis send audio anywhere?
No. Analysis is offline, background and fully local. No cloud, no account, ever.
Can I track through it?
Yes. Track mode runs zero-lookahead dynamics on the Classic tune path with 4 samples of total latency at 48 kHz, 0.1 ms, test-enforced under a 5 ms contract. Mix mode reports its latency exactly in the header.
Can I use it on a group or a polyphonic part?
Burnish is a lead-vocal chain: mono or stereo in, stereo out. Polyphonic material is out of scope by design.
Which formats and platforms?
VST3, AU and AAX on Windows 10+ and macOS, Intel and Apple Silicon. That covers Pro Tools, Logic Pro, Ableton Live, Cubase, Studio One, FL Studio, REAPER and every other major host.
Is there a free trial?
Yes. Every released Vitric plugin has a free 3 day trial with every feature available during the trial window. Get the installer and your machine-bound trial license from the downloads page. At the signed expiry, audio processing and generated MIDI lock until you install a purchased license or a valid subscription grant. The plugin can still load so projects and saved settings are not deleted.
More from the line
Burnish Vocal Chain
Sits like the record. Never sounds identical to it.
Eight stages, eight scopes, one window, and a reference that explains itself.