VocalsRecording

Recording Vocals In A Room You Cannot Treat

Reflections are a distance problem, not a gear problem. What to change at the microphone in an untreated room, and which damage no later pass can undo.

All articles

The Room Is A Distance Problem

What makes an untreated room audible in a vocal take is not the price of the room, it is the ratio between the sound that reaches the microphone directly and the sound that reaches it after bouncing. Direct sound falls away quickly with distance while reflected sound arrives from every direction at roughly the same level, so the decisions that matter all happen before the take: where you stand, how close you get, and what the microphone is pointed into. Repair afterwards is a real option for part of this, and only for part of it.

What To Change Before You Record

Get Out Of The Middle Of The Room

The centre is the worst place to stand because it sits equidistant from every pair of parallel surfaces, so reflections arrive together and reinforce each other instead of spreading out in time. Move off centre, stand at an angle to the walls, and avoid facing a flat surface square on. None of this costs anything and it changes the recording more than most things you could buy.

Move Closer, And Pay The Proximity Bill

Halving the distance to the microphone raises the direct sound without raising the room, which is the only free improvement available in the whole chain. The bill arrives as proximity boom, plosives and every mouth noise you own. A hand span rather than an arm's length, slightly off axis so breath passes the capsule instead of hitting it, is the trade most untreated rooms want.

Put The Soft Surface Behind The Microphone

The strongest single reflection in a vocal take is usually the one that leaves the singer, travels past the microphone, hits whatever is in front of the singer, and comes back into the most sensitive side of the capsule. That surface is where a duvet belongs. Padding wrapped around the back of the microphone covers a direction it already rejects, which is why it changes so little.

Break Up Any Two Facing Hard Surfaces

Flutter echo needs a pair of parallel reflective walls to bounce between, and it is the artifact that survives everything later because it is fast, repetitive and tonal. Softening one of the two surfaces is enough to stop it. If you cannot soften either one, stand so that the path running between them does not pass through the microphone.

Level Settings Do Not Change The Room

Gain multiplies the direct sound and the room by exactly the same amount, so recording quieter does not reduce reflections, it only moves both further down toward the noise floor. Singing louder does not help either, for the same reason. Three things actually move the ratio: the distance between mouth and capsule, how much of the room gets absorbed before it arrives, and which directions the microphone and the voice are throwing energy in.

Directivity is the part people forget. A cardioid pattern rejects what is behind it, but room sound arrives from everywhere, including the front, so the pattern buys you less than the specification suggests. Your own voice is directional too, and it is most directional at high frequencies. Turning a few degrees away from the hardest wall changes what that wall gets fed, which changes what it sends back, and it costs one step.

What Can Be Repaired, And What Cannot

A model trained on room decay can recognise a tail: the diffuse, delayed copy of your voice that keeps ringing after the word has ended. A moderate tail is genuinely removable, and taking that decay off a finished vocal is a normal step now rather than a rescue operation. The offline cleanup in RAW runs on the file after the take, which means the part of the room that is separable in time from the source can come out.

Comb filtering is the part that cannot. When a hard surface close to the singer returns a copy within a few milliseconds, that copy lands inside the word rather than after it and sums with the direct sound, cancelling some frequencies and boosting others. What you have is a filter, not a tail, and it is now part of the timbre. It also shifts whenever the singer moves, so no fixed EQ curve in the vocal chain can correct it later.

Signs The Room Got Into The Take

  • Words ring on after they end and the ring holds one pitch regardless of the note, because you are hearing the room's own modes decay.
  • Moving your head changes the tone more than it changes the level, because a nearby hard surface is combing with the direct sound.
  • The take turns thin and hollow when you sum it with a double, because a close reflection already filtered one of the two.
  • Everything gets worse the moment you step back, because distance always favours the room over the source.
  • The vocal sits behind the beat no matter how hard you compress it, because the tail is filling the gaps compression is trying to open.
  • Turning the preamp down cleans nothing, because gain scales the room and the voice by the same amount.

For The Room You Kept

RAW takes the decay off a finished take offline: select the range on the waveform, then drag the dry version straight into the project.

See RAW

Guides

You may also like

Our latest news

More guides