Microphone Settings Explained: Gain, Sample Rate and Bit Depth

Last updated 8 August 2026 — MicTest.live

Open any microphone's properties and you are presented with numbers that sound technical and rarely come with an explanation. Sample rate, bit depth, gain, three kinds of processing. Most guides tell you to set them higher. That is usually wrong, and occasionally harmful. Here is what each one really does.

Sample rate

Sample rate is how many times per second the microphone's analogue signal is measured. 44,100 samples per second — 44.1 kHz — is the CD standard; 48 kHz is the video and broadcast standard and the default in most operating systems.

The reason those numbers exist is a mathematical result: to capture a frequency accurately you need to sample at more than twice that frequency. Human hearing tops out around 20 kHz, so a little over 40 kHz captures everything anyone can hear. That is the entire justification for 44.1 kHz, and it is sufficient.

Higher rates — 96 kHz, 192 kHz — capture frequencies nobody can hear, at the cost of double or quadruple the file size and processing load. They have genuine uses in production, mainly because slowing audio down for sound design preserves more detail. For a voice call or a podcast they change nothing you can perceive.

Bit depth

If sample rate is how often the signal is measured, bit depth is how precisely each measurement is recorded. 16-bit gives 65,536 possible levels; 24-bit gives about 16.7 million.

What that buys you is dynamic range — the distance between the quietest sound you can capture and the loudest before clipping. 16-bit provides around 96 dB, which comfortably exceeds what any room and microphone will deliver. 24-bit provides around 144 dB.

The practical benefit of 24-bit is not better sound at a correct level; it is headroom for mistakes. Record too quietly at 16-bit and raising the level afterwards brings up audible noise. Do the same at 24-bit and there is enough resolution below your signal that the recovery is clean. That is why studios record at 24-bit and deliver at 16-bit.

Gain

Gain is the amount of amplification applied to the microphone signal before it is converted to digital, and it is the single most important setting on this page. Everything downstream is working with whatever gain captured.

Too little gain and your voice sits close to the noise floor; raising it afterwards raises the noise with it. Too much and the loudest peaks exceed what the converter can represent, and those peaks are clipped flat — an unrecoverable distortion that sounds harsh and crackly.

The target is peaks around −12 to −6 dBFS with an average near −20 dBFS. That leaves room for a sudden laugh or a raised voice without clipping, while staying well above the noise floor. The microphone test shows all three numbers live so you can set this by eye in under a minute.

Echo cancellation

Acoustic echo cancellation removes sound that came from your own speakers from your microphone signal. The software knows what it just played, so when that signal reappears at the microphone a few milliseconds later it subtracts it.

It exists for speakerphone situations, and there it is essential. It is unnecessary when you wear headphones, because there is no path from speaker to microphone. It also has a cost: the subtraction is imperfect and can slightly dull a voice, which is why it is switched off for recording.

Noise suppression

Noise suppression identifies steady background sound — fans, air conditioning, traffic hum — and removes it. Modern implementations use machine learning models trained to separate speech from everything else, and they are impressively good at constant noise.

They are much weaker at sudden noise, because a door slam looks nothing like the steady profile they learned. And they have a characteristic failure mode: when your voice is quiet, the model can classify it as noise and remove it, so listeners hear you cutting in and out. If people say you keep dropping mid-sentence, this is the first thing to turn off.

Automatic gain control

AGC continuously adjusts the input level to keep you at a consistent loudness. It rescues people who move around while talking or who vary a lot in volume.

The cost is that it raises the level during quiet passages, which raises the noise floor with it. You hear this as the room hiss swelling up between sentences and ducking when you speak — a distinctive pumping effect. Studio recordings almost always have AGC off with the gain set manually, because a consistent noise floor matters more than consistent loudness in material that will be edited.

What to actually use

Setting	Video calls	Podcast / voiceover	Music
Sample rate	48 kHz	48 kHz	48 kHz (96 kHz if your workflow needs it)
Bit depth	16-bit is fine	24-bit	24-bit
Gain	Peaks around −12 dB	Peaks around −6 dB	Peaks around −6 dB
Echo cancellation	On (off with headphones)	Off	Off
Noise suppression	On	Off — treat the room instead	Off
Auto gain control	On	Off	Off

You can hear each of these decisions rather than take them on faith. The microphone test lets you toggle all three processing options while watching the meter and recording short takes, so the difference between processed and raw is immediately obvious.

More guides