The usual fix puts a compressor after the reverb on its own bus, with the sidechain input keyed to the dry vocal rather than the wet. Common starting points run around 10 to 30 milliseconds of attack and 100 to 300 of release, with a moderate ratio. It works, and it has worked for a long time.
It also solves one narrow version of the problem. Most people never notice the parts it leaves alone, because those parts sound like the reverb being slightly disappointing rather than like a technique failing.
Start with what the compressor is actually doing. It reads one number, the level of the key signal, and it pulls the entire reverb output down by an amount tied to that number. Nothing more.
A standard compressor does not care. It drops 200 Hz and 8 kHz by the same amount, so every time it deals with the mud it also strips the air off the tail that made you reach for reverb in the first place.
You can chase this with a multiband compressor or a dynamic EQ on the return, and plenty of people do. Now you are managing crossover points and several sets of attack and release, all to make one reverb get out of the way.
AURA splits the job instead. Its clarity control handles the whole tail while duck lows pulls the bottom down harder on its own attack and release, so the mud that piles up under a busy chorus gets dealt with without gutting the top.
Your compressor keys off the lead vocal. When the lead sings, the reverb ducks for the backing vocals, the guitars, and the piano too, because all of them share the bus. Everything in the mix starts breathing in time with one performance.
The workaround is a dedicated send and a dedicated compressor for every source that needs one. Four sources means four reverb instances, four compressors, and four sets of routing to keep straight.
Running the ducker inside the reverb collapses all of that. Each instance handles the signal passing through it, and a reverb that ducks itself needs no key routing at all unless you deliberately want an outside trigger.
It cannot tell that you moved from an F to a G, so the old note keeps ringing inside the tail underneath the new one. That collision is what turns sustained chords and legato vocal lines to soup, and no threshold setting reaches it.
AURA's tune control works on the pitch content rather than the level, clearing the tail between the notes you are actually playing. Drain sits alongside it and sets how hard old notes get pushed out as new ones arrive.
The gain reduction readout matters here too. On a bus compressor, you are watching one meter for the entire return, so you cannot tell whether the duck is landing on the source you care about or on something else that happened to be loud in the same bar.
None of this makes the bus method wrong. Plenty of experienced mixers argue that ducking is a special-case tool anyway, and that arrangement fixes more mud than any sidechain ever will. That holds right up until someone hands you a dense track you did not arrange, with a vocal that has to sit on top of it and a deadline that rules out rebuilding the parts.
It also solves one narrow version of the problem. Most people never notice the parts it leaves alone, because those parts sound like the reverb being slightly disappointing rather than like a technique failing.
Start with what the compressor is actually doing. It reads one number, the level of the key signal, and it pulls the entire reverb output down by an amount tied to that number. Nothing more.
A broadband duck flattens the part you wanted
Reverb mud lives mostly at the bottom. A long tail under a vocal piles up somewhere below 300 Hz, and that is the range smothering the words.A standard compressor does not care. It drops 200 Hz and 8 kHz by the same amount, so every time it deals with the mud it also strips the air off the tail that made you reach for reverb in the first place.
You can chase this with a multiband compressor or a dynamic EQ on the return, and plenty of people do. Now you are managing crossover points and several sets of attack and release, all to make one reverb get out of the way.
AURA splits the job instead. Its clarity control handles the whole tail while duck lows pulls the bottom down harder on its own attack and release, so the mud that piles up under a busy chorus gets dealt with without gutting the top.
A shared reverb bus ducks the wrong tracks
Most sessions run one or two reverbs for a dozen sources. That is good practice for glue, and it breaks ducking outright.Your compressor keys off the lead vocal. When the lead sings, the reverb ducks for the backing vocals, the guitars, and the piano too, because all of them share the bus. Everything in the mix starts breathing in time with one performance.
The workaround is a dedicated send and a dedicated compressor for every source that needs one. Four sources means four reverb instances, four compressors, and four sets of routing to keep straight.
Running the ducker inside the reverb collapses all of that. Each instance handles the signal passing through it, and a reverb that ducks itself needs no key routing at all unless you deliberately want an outside trigger.
Level detection cannot hear which note you played
The deepest limit has nothing to do with routing. A compressor knows how loud the input is, and that is the whole of what it knows.It cannot tell that you moved from an F to a G, so the old note keeps ringing inside the tail underneath the new one. That collision is what turns sustained chords and legato vocal lines to soup, and no threshold setting reaches it.
AURA's tune control works on the pitch content rather than the level, clearing the tail between the notes you are actually playing. Drain sits alongside it and sets how hard old notes get pushed out as new ones arrive.
The gain reduction readout matters here too. On a bus compressor, you are watching one meter for the entire return, so you cannot tell whether the duck is landing on the source you care about or on something else that happened to be loud in the same bar.
None of this makes the bus method wrong. Plenty of experienced mixers argue that ducking is a special-case tool anyway, and that arrangement fixes more mud than any sidechain ever will. That holds right up until someone hands you a dense track you did not arrange, with a vocal that has to sit on top of it and a deadline that rules out rebuilding the parts.
