Microphone Arrays Explained: How Multiple Mics Work Together

Odds are you’re using a microphone array right now without realizing it. Your laptop has one, your phone has one, and that smart speaker on your kitchen counter?

Definitely has one.

Here’s the thing though: a microphone array isn’t just “a bunch of mics.” It’s a deliberate arrangement of identical microphones wired into a signal processor, and together they pull off stuff a single mic physically can’t.

Think focusing on one voice in a loud room, figuring out which direction a sound came from, or killing echo on a video call.

What Is a Microphone Array?

Strip away the jargon and a microphone array is two or more mics set up in a specific pattern and wired to a digital signal processor (DSP). The DSP grabs the output from all of them and analyzes it as one combined signal.

One detail that trips people up: the mics have to be identical. Same type, same sensitivity, same polar pattern.

Mix a condenser with a dynamic, or pair two mics with different gain levels, and the whole system falls apart because the math stops working.

That’s also what separates an array from, say, miking a drum kit with five different mics. On the kit, each mic handles its own job independently.

In an array, the signals get processed together, and the DSP compares the differences between them to extract information no single mic could ever give you on its own.

Like what?

• Direction of arrival: Which direction is the sound coming from?
• Source separation: Can we isolate one voice from background noise?
• Spatial mapping: Can we build a picture of where every sound source is in the room?

How Does a Microphone Array Work?

Sound travels through air at roughly 343 meters per second, which sounds fast but isn’t instant. When a sound wave crosses a row of microphones, it hits the nearest one first and the farthest one last.

The time gap is absurdly small, just microseconds, but it’s predictable.

If the sound is coming from the left, the leftmost mic picks it up a hair before the rightmost one. The DSP measures those arrival-time differences and works backward to figure out exactly where the sound originated.

Once it knows the direction, it can do something really useful: boost the signals coming from that direction and squash everything else. That’s beamforming, and it’s basically the whole reason microphone arrays exist.

But direction-finding isn’t the only trick. Because the array has independent signals from multiple positions, the DSP can also subtract ambient noise that hits all mics equally (like HVAC hum) from the focused signal.

This is how arrays achieve noise cancellation that goes way beyond what any single mic’s polar pattern can manage.

Beamforming: The Core Technology

Beamforming is where microphone arrays really start to shine. It’s a signal processing trick where the DSP applies tiny time delays to each mic’s signal before adding them all together.

The sequence goes roughly like this:

1. Sound hits each mic at a slightly different moment
2. The DSP time-shifts each signal so that sound from the direction you care about lines up perfectly across all channels
3. Those aligned signals add up and get louder (constructive interference)
4. Sound from other directions stays misaligned, so it partially cancels itself out (destructive interference)

What you end up with is a virtual “beam” of sensitivity aimed at your target, with everything off to the sides getting quieter. Hence the name.

Two flavors are worth knowing about:

Delay-and-sum is the basic version. The processor applies fixed delays based on the array’s geometry and where you want to listen.

It’s computationally lightweight and does the job for straightforward setups like conference room ceiling microphones.

Adaptive beamforming is the smarter cousin. Instead of fixed delays, the processor constantly recalculates based on what it’s actually hearing, tracking moving sources and adjusting for changing noise.

Smart speakers like the Amazon Echo use this approach, which is how they follow your voice as you wander from the living room to the kitchen.

Types of Microphone Array Configurations

Geometry matters more than most people expect. How you physically arrange the mics dictates what the array can and can’t do.

Linear Arrays

Mics in a straight line. These are great for figuring out if a sound is coming from the left or right, which is why you’ll find them in soundbars and along laptop screen bezels.

Dead simple to build, but they can’t tell you if a sound is above or below.

Circular Arrays

Mics arranged in a ring. This is what most smart speakers use (Amazon Echo has 7 in a circle, Google Nest has 3).

A circular layout gives you 360-degree awareness on the horizontal plane. It’s why your Echo can tell whether you’re talking from the kitchen or the hallway.

Planar Arrays

Mics spread across a flat grid. These resolve direction in two dimensions and are used in acoustic cameras, the kind that generate visual “heat maps” of where sound is coming from.

The Dutch company Sorama holds the record with a 4,096-MEMS-microphone planar array.

Volumetric Arrays

Mics positioned in three dimensions, think a sphere or a cube. Full spatial resolution.

You’ll mostly see these in research labs, VR audio capture rigs, and military acoustic surveillance systems.

End-Fire vs. Broadside

These describe how the array faces the sound. Broadside means sound arrives perpendicular to the mic line.

End-fire means sound comes along the axis. End-fire setups produce tighter beams, but the timing has to be more precise or it falls apart.

Microphone Arrays vs Single Microphones

Even the best single mic has hard limits that an array blows past:

Capability	Single Mic	Microphone Array
Noise rejection	Limited to polar pattern	Active suppression via beamforming
Sound localization	Cannot determine direction	Pinpoints source direction and distance
Echo cancellation	Not possible alone	Built-in acoustic echo cancellation (AEC)
Far-field pickup	Degrades rapidly with distance	Maintains clarity across a room
Adaptability	Fixed response	Can steer beam dynamically

A directional microphone like a shotgun can reject off-axis noise through its physical design, sure. But it can’t adapt on the fly or electronically steer where it’s pointing.

An array can.

The trade-off is complexity. More mics, more data to crunch, more hardware that needs to match, more cost.

If you’re close-miking a vocalist in a treated studio, a single good condenser microphone will beat an array. No question.

Arrays earn their keep in the tough environments: noisy rooms, long distances, echo-heavy spaces.

Where Microphone Arrays Are Used

Smart Speakers and Voice Assistants

The Amazon Echo packs 7 mics in a circular array. Google Nest uses 3.

Apple’s HomePod has 6. That’s how these things pick up your voice from across the room even when music is blasting.

Far-field voice recognition wouldn’t exist without arrays.

Video Conferencing

Conference room systems from Poly, Shure (their MXA920 ceiling array uses up to 113 tiny mics), and Sennheiser all rely on ceiling microphones with array tech baked in. They pick up every person at the table while filtering out HVAC drone and room reverb.

Most of these also handle acoustic echo cancellation (AEC), which prevents the far-end caller’s voice from feeding back through the room speakers and into the mics. Even your laptop’s webcam probably has 2 to 4 MEMS mics lined up along the bezel doing the same thing on a smaller scale.

Smartphones

Your phone likely has 2 or 3 MEMS (Micro-Electro-Mechanical Systems) mics working as an array. These are tiny silicon-chip microphones, basically microscopic mechanical structures etched onto a semiconductor wafer.

One grabs your voice, the others sample the ambient noise around you, and the DSP subtracts one from the other. That’s why the person on the other end can hear you clearly even on a windy street.

Hearing Aids

Beamforming microphones in modern hearing aids use a two-mic array to focus on the person in front of the wearer while suppressing crowd noise from the sides and rear.

Automotive

Your car’s hands-free calling system uses an array mounted somewhere in the headliner or near the steering column. It locks onto the driver’s voice and filters out road noise, wind, and the kids arguing in the back seat.

Acoustic Imaging

Engineers use large planar arrays (dozens to thousands of mics) to create visual maps of sound. This is used to find rattles in car interiors, locate noise sources in machinery, and optimize building acoustics.

Military and Surveillance

Arrays of long-distance microphones are used for acoustic detection and ranging. By measuring the time differences across a wide-baseline array, military systems can determine the direction and distance of gunfire, aircraft, or vehicles.

Key Performance Metrics

Whether you’re shopping for a conference array or building your own, four numbers tell you most of what you need to know:

Signal-to-Noise Ratio (SNR) Gain is how much cleaner the array’s output is compared to a single mic. Two mics get you about 3 dB of improvement.

Double the mic count again and you gain another 3 dB. It adds up.

Directivity Index (DI) tells you how tight the beam is. Higher DI, narrower focus.

More mics and wider spacing both push the DI up.

Array Factor (AF) maps out the combined pickup pattern. It shows you where the main lobe is pointing (your target) and where the side lobes are (directions you’d rather ignore).

Frequency Response is where things get tricky. Low frequencies are hard to focus because their wavelengths are long, and you’d need impossibly wide mic spacing to wrangle them.

High frequencies steer easily, but if your mics are too far apart, you get “grating lobes,” basically phantom beams pointing in wrong directions.

Rule of thumb: keep the mic spacing under half the wavelength of your highest target frequency. For speech frequencies (up to about 8 kHz), that works out to roughly 2 centimeters between mics.

Building a DIY Microphone Array

Good news: you don’t need 4,096 mics to get started. Two mics, a stereo interface, and a laptop are enough to hear beamforming working with your own ears.

What You Need

• Two identical microphones (same model, same brand). Omnidirectional electret condensers work well.
• A stereo audio interface or recorder that can capture both channels simultaneously
• A computer with recording software (Audacity works fine)
• A rigid mounting bracket to hold both mics at a fixed distance

Assembly Rules

Match everything. Same exact mic model, same sensitivity, same directionality.

Professional arrays spec this tightly: sensitivity matched within ±1.5 dB and phase matched within ±1.5 degrees. You don’t need lab-grade precision for a home project, but even a small mismatch will noticeably wreck the noise rejection.

Lock down the spacing. Mount them at a fixed distance.

For speech, somewhere between 2 and 10 centimeters works well. Go wider for better bass response, but you’ll lose some high-frequency accuracy.

Make it rigid. The mics cannot shift relative to each other.

Even a slight wobble corrupts the phase relationship between the two channels. A solid mounting bar, or a 3D-printed bracket if you’re handy, does the trick.

Record both channels at once. No exceptions.

Both tracks have to be captured simultaneously through the same interface. Two separate recorders with their own clocks won’t cut it.

Once you’ve got your two synced tracks, open them in Audacity and try applying a small time delay to one channel before summing them. You’ll hear it immediately.

Sound from the direction you’re targeting gets louder, and the off-axis stuff fades back. That’s beamforming, and you just did it with two mics and free software.

Frequently Asked Questions

How many microphones do you need to build a microphone array?

You need at least two microphones of the same type working simultaneously and connected to a processor. There’s no upper limit.

Professional arrays can include dozens or even thousands of mics depending on the application. The Dutch company Sorama built one with 4,096 individual microphones.

Are microphone arrays the same thing as using multiple mics on stage?

No. A microphone array is a coordinated system where multiple mics work together and their signals are processed as a group for beamforming, noise cancellation, or spatial audio. Simply using several separate mics on stage for different instruments is not an array because the signals are not jointly processed.

Do smart speakers like Amazon Echo use microphone arrays?

Yes. Most smart speakers use circular microphone arrays with 4 to 7 MEMS mics arranged around the top of the device.

This lets them detect which direction your voice is coming from, focus on your speech, and filter out background noise through beamforming.

What is beamforming in a microphone array?

Beamforming is a signal processing technique that combines the output of multiple microphones to focus on sound from a specific direction. It works by analyzing the time delays between when sound arrives at each mic, then amplifying signals from the target direction while canceling noise from other directions.

Can a microphone array cancel background noise?

Yes. Because the array can distinguish between sounds arriving from different directions, it can suppress noise coming from everywhere except the target direction.

This is why conference room arrays pick up the speaker’s voice clearly even in a noisy open-plan office.

Final Thoughts

Microphone arrays quietly power a huge chunk of the devices we use every day. Every time you talk to Alexa, hop on a Zoom call, or dictate a text message, an array is working behind the scenes to keep your voice clean.

The core concept is dead simple: multiple identical mics, arranged in a pattern, with a processor that analyzes the time differences between them. From that one idea you get beamforming, noise cancellation, echo suppression, and sound source localization.

If you decide to build your own, matching is everything. Same mic model, same sensitivity, same directionality.

Get those right and even a basic two-mic array will out-punch a single mic in noisy environments.