How far should a microphone be from your mouth?

You bought a microphone because the sound was bad, and the sound is still bad. It is bad differently now, which is a kind of progress, but it is bad.

Published distance ranges are useful starting points, but the microphone, voice and room decide the position. Too close can create plosives or overload; too far can bring in more room sound. Test both before the main recording.

Why is there no correct number?

Because the distance is a trade, and both ends cost you something.

Bring it nearer and you get more of your voice relative to the room, which is what you want, and more of everything your mouth does that is not speech, which is not. Move it away and the mouth noise recedes, and the room comes forward to take its place.

There is no position where both are perfect. There is a position where both are acceptable, and it sits somewhere different for a softly spoken person in a carpeted study than for a loud one in a tiled kitchen. A page giving you a single number has decided which of those you are.

What goes wrong when you are too close?

Three things, in roughly ascending order of how much they cost you.

Plosives

A plosive is the burst of air that leaves your mouth on a P or a B. It is not a loud sound, it is a physical puff, and when it hits the microphone directly the recording gets a low thump underneath a word that had no thump in it. A foam cover helps and is about the cheapest fix in audio. Angling the microphone off the centre line of your mouth helps more, because the puff travels forwards while your voice travels everywhere.

The s sounds

Move close to most microphones and your s and t sounds sharpen, then start to hiss. It can be reduced afterwards, up to a point, and past that point the treatment begins to lisp the voice. Far better not to record it.

Clipping

Clipping is what happens when a signal is pushed louder than something in the chain can describe: the microphone capsule, the preamp, or the recorder's input. It flattens the tops off the loudest moments and gives them back as a crackle, and it is permanent, because nothing was written down to restore. The usual cause is gain staging, the level set too high for how loud the source really is, so the fix is almost always to turn the input gain down and leave headroom rather than to move. Distance feeds into it, since being closer arrives hotter and a level that was fine at a hand's span can clip when you lean in or laugh partway through a take. Set the gain for the loudest thing you are going to do.

What goes wrong when you are too far away?

More room sound relative to the direct voice.

The room arrives. Every hard surface sends a copy of your voice back to the microphone slightly late, and as you move away the direct sound gets quieter while the reflected sound stays where it was. The proportion tips. The recording stops sounding like you in a room and starts sounding like a room with you somewhere inside it.

That is the one fault clean-up cannot work around. A hum can be reduced. A siren can sometimes be cut out. Reflections are made of your own voice, arriving late, occupying every frequency the thing you want to keep occupies, so removing them removes part of you. Software tries, and it is far better than it was: the current tools pull a roomy recording back to usable, where the older ones handed back something thin and underwater. It is still reconstruction, it still costs a little of the voice, and heavy reflections still win, which is why distance comes first.

A starting point, not a rule

For a microphone you speak into rather than one you aim, roughly a hand's span from your mouth is a sensible place to begin: somewhere between fifteen and thirty centimetres, positioned slightly off the centre line so you are not breathing straight down the barrel.

That is a starting point and it is marked as one deliberately. A shotgun microphone, the long thin kind that gets pointed at somebody rather than spoken into, works at a different distance altogether, usually further away and above, just out of frame. A clip microphone on your chest has its own rules. And your own voice overrules all of it: speak softly and you will end up nearer than the range, project and you will end up further away.

Should the microphone be in front of your mouth or off to one side?

Off to one side, slightly, and slightly above or below your mouth rather than dead level with it.

The reason is the plosive again. That puff of air travels forward in a narrow cone, so a microphone just outside the cone hears the same voice and none of the air. It also stays out of frame more easily, and you stop drifting closer to it, because nothing is in front of your face inviting you to lean in.

Consistency matters more than precision. Most directional microphones, which is to say the cardioids most people film with, get noticeably bassier as you move in close; it is called the proximity effect, and an omnidirectional microphone barely shows it. That is why somebody who leans forward on the interesting bit can sound like two different people in one video. Pick a position, mark it, hold it.

Your own number

Record yourself three times, and listen back on headphones; laptop speakers hide the faults you are listening for.

Say the same twenty seconds at three distances: as close as you can bear, a hand's span away, and an arm's length. Include a sentence with several p and s sounds in it, and one where you laugh, because that is where the failures live rather than in calm speech.

Then listen to all three in a row, checking two things. Can you hear the room. Does anything bump or hiss. Take the closest position where nothing bumps or hisses, and if you can still hear the room from there, distance is not the problem you are solving and there is a separate post for what is.

Mark the useful position and keep a reference recording. Check it again when the microphone, room, speaker or recording task changes.

Sources and evidence

What here is established acoustics and what is our own method.

What is established. Closer is not simply better: as a microphone comes in, its low end lifts through the proximity effect, which depends on the microphone's polar pattern and is documented by DPA Microphones, while moving out trades that for more of the room, the direct-to-reverberant balance the Institute of Professional Sound calls critical distance. No single distance is right for every microphone and voice.

Our method. Setting the distance by ear for the particular microphone and voice, close enough to beat the room but clear of the breath, is the practice here.

References.

• DPA Microphones. Proximity effect in microphones explained. dpamicrophones.com. Distance response depends on microphone design and polar pattern. Manufacturer education.
• Institute of Professional Sound. Critical distance. ips.org.uk. Room sound depends on direct-to-reverberant balance, not distance alone. Technical reference, not peer-reviewed research.