The S cuts through the whole mix
Pull it back and the vocal goes dull. Leave it and every “s” is the loudest thing in the record. You fix one line and the chorus bites again.
Your vocal, without the stack.
The Chemistry listens to 30 seconds of your take, works out what is masking it, and sets the chain. One insert, VST3 or AU on macOS.
No payment now. One email when the founding licences open: $99, then $129. The first hundred only.
You can hear that it is not there yet.
The vocal sits on top, not inside
Turn it up and it covers the drums. Turn it down and the words disappear. You nudge it for twenty minutes and it never quite lands.
Ten plugins and no idea which one helped
Gate, EQ, compressor, de-esser, another EQ, saturation, a reverb send, a limiter. The chain became a second job, and you still are not sure.
Three takes, one chain.
The role selector changes what the chain does, not just what the preset is called.
Not for you if you work in Pro Tools, you need Windows, or you are looking for a tuner. Those are real limits, and they are better said now than after you pay.
You sing, and you mix yourself
Music role on the lead. It takes the weight out of the low mid, holds the sibilance down and puts the take in front of the beat without turning it into a different voice.
You produce for other people
Music on the lead, Bus across the stack. One Listen gets the rough to the point where the artist will actually sing over it, and the Listen Report shows what moved.
You record voice for a living
Dialogue role. Podcasts, narration and voiceover: it removes the room and the plosives and leaves the consonants intact, which is where most dialogue chains go wrong.
One
It listens to 30 seconds.
Playback stays dry and audible. It traces the input, the cuts and the boosts, and refines at 35, 65 and 88 percent while your transport keeps running.
Two
It takes the harshness out, band by band.
HARSH at 2 to 5 kHz, SIB at 5 to 9, AIR above 9. Each band has its own gain, so a 3 kHz consonant cannot clamp the 12 kHz air.
Three
It sets the rest of the chain in order.
Noise, clean, tone, dynamics, character, glue. Every macro it moved stays on its own card, so you can see the move and argue with it.
Four
You keep the last word.
RAW or READY, level-matched, so you are not fooled by loudness. Revert it, save it as a preset, or export the Listen Report as the proof.
What it replaces
One insert does what a stack of plugins was supposed to do. The gain rider, the gate, the de-esser, the dynamic EQ, the room, the glue and the true-peak limiter live in one window, in signal order, with the settings visible.
$99
One payment. No subscription, no cloud account, no phone home. It runs on your machine and it keeps working when the network does not.
If it is not clearer after one pass, ask for the refund inside 30 days and keep every preset you saved.
In the founding licence
- The Chemistry for macOS, VST3 and AU, Apple Silicon and Intel
- A 30 second Listen with Music, Dialogue and Bus roles
- Three de-ess bands and an adaptive five-band dynamic EQ
- Character, room, glue and a true-peak limiter
- A shareable Listen Report, as PNG and JSON
- One year of updates
Try it on a take you already gave up on.
Load the raw vocal you could not finish. Run one Listen. If the result is not clearer and more balanced, ask for a refund within 30 days. You keep the result and every setting you saved, and the licence goes back into the first hundred.
Questions, answered plainly.
Is this an AI plugin?
No cloud model and no training data. The Listen engine is signal analysis and fixed rules, so the same take gives the same result every time. Nothing is uploaded, and there is no account to create.
Will it work on my voice?
That is what the roles are for. Music for sung vocals, Dialogue for speech, Bus for a group or the whole mix. If it does not work on your take, the refund is there and you keep the settings.
Does it tune?
No. The Chemistry is a channel: it cleans, balances, de-esses, adds depth and limits. Tuning happens before it, in whatever tuner you already trust.
Windows, or Pro Tools?
macOS only for now, as VST3 and AU. Apple Silicon and Intel. Windows and AAX come later, and only if enough people ask.
What if I hate what it did?
Press Revert and the pre-Listen settings come back. Every macro is yours to move afterwards, and you can A/B against the raw at matched loudness before you commit to it.
Does it phone home?
No. There is no licence check against a server, no telemetry and no crash reporting. Once it is on the machine it works offline.
The first hundred get it at $99.
After that the price is $129 and it does not come back down. When the founding window closes, the next email you get is the one with the licence in it.
Leave your address and you will hear before anyone else does. No payment now, and nothing to install until it opens.
Reserve a founding licence$99 founding · $129 after · 30-day refund · macOS only