What Is Microphone Peaking? Causes, Fixes, and How to Prevent It

Do your recordings sound distorted or crackly? Well, that's your mic hitting peak. It is common among podcasters, streamers, and creators. This happens when the signal gets pushed too hard and starts breaking apart. The result is harsh distortion that can ruin a take. Once you understand where it starts, controlling it becomes easy. You can deal with it at different points in your setup, from the mic itself to your recording software.

What Is Microphone Peaking?

Microphone peaking happens when an audio signal exceeds the maximum recordable digital level, known as 0 dBFS (decibels relative to full scale). Once the signal passes that limit, the recording system cannot hold the extra loudness. The tops of the waveform are sliced off flat, producing what is called hard clipping.

Peaking happens when the audio level reaches the top limit on a meter, usually 0 dBFS or higher. Clipping is what you hear after that point. It shows up as sharp, broken distortion when the waveform gets cut off. Peaking does not always cause noticeable clipping. A brief, isolated peak may pass without obvious damage. But sustained or repeated peaks almost always produce audible distortion.

The problem can originate at multiple points in the signal chain:

• The microphone capsule itself, if it is overloaded by loud sound pressure

• The transmitter gain stage on a wireless microphone

• The preamp on an audio interface

• The input level set inside recording software

Understanding where the peak originates is what determines the right fix, which is why the solutions below work through each stage in order.

What Causes Microphone Peaking?

Peaking almost always comes down to too much signal entering a stage that cannot handle it. These are the four most common reasons that happen.

Input Gain Set Too High

Gain is the amplification applied to the microphone signal before it reaches the recording stage. When gain is pushed higher than the source loudness requires, the amplified signal can exceed 0 dBFS before the recording software even registers the problem. This is the most frequent cause of peaking across every microphone type, whether wired, USB, or wireless. The fix almost always starts here.

Loud or Unpredictable Sound Sources

Even properly set gain can still get overloaded by sudden loud sounds. A presenter who shouts a punchline, a burst of laughter during a podcast interview, or the attack transient from a snare drum can spike 10 to 15 dB above the average level the gain was calibrated for. A gain setting dialed in for normal conversational speech will clip the moment the source gets unexpectedly loud.

Microphone Placed Too Close to the Source

Sound pressure level increases significantly as the distance between source and capsule decreases. Placing a microphone just 2 to 3 inches from a speaker’s mouth or a musical instrument raises the SPL at the capsule dramatically. Even a moderate gain setting can cause clipping at close range because the capsule receives more acoustic energy than the gain structure is designed to handle.

High-Sensitivity Capsule in a Loud Environment

Some microphones are built specifically for quiet recording environments such as voiceover booths, broadcast studios, or desktop use. Their capsules are highly sensitive by design, converting sound to electrical signal very efficiently. When these microphones are used outdoors, on a live set, in a performance environment, or close to a loud speaker, the capsule can overload even before any gain amplification is applied. In these cases, a hardware pad or a lower-sensitivity microphone is the more appropriate solution.

How to Recognize Microphone Peaking?

Before adjusting anything, confirm that peaking is the actual problem rather than another audio issue. Here is what to look for:

• Red or orange clip indicators on your meters: Recording applications such as Audacity, OBS, GarageBand, and most DAWs display clip indicators that illuminate when the signal reaches or exceeds 0 dBFS. A brief flash is a warning. A lit indicator that stays on is a problem that needs immediate attention.

• Flattened waveform peaks: Open the recorded audio in a waveform editor. Hard-clipped audio shows the tops and bottoms of the waveform cut off flat rather than forming natural rounded peaks. This visual cue is one of the clearest signs of clipping.

• Crackling, buzzing, or harsh distortion on playback: Clipping has a sharp, harsh crackling sound. It is different from soft compression or tape-like saturation. It usually feels unpleasant and tiring to listen to.

• Consistent versus occasional peaking: If the meters are permanently in the red, the gain is set too high across the board. If the clip indicator only flashes at specific moments, source dynamics are exceeding an otherwise acceptable gain level. Both situations need fixing, but the diagnosis points toward different solutions.

How to Fix Microphone Peaking?

Work through these steps in order. Each one addresses a different point in the signal chain, moving from the audio source outward toward the software.

Step 1: Reduce Input Gain at the Source

The most effective fix is always to reduce gain at the earliest point in the signal chain. On an audio interface, this means turning down the preamp gain knob. On a wireless microphone system, it means adjusting gain at the transmitter or receiver, not only at the software input level downstream.

This difference matters a lot for wireless microphone users. Lowering the level in software like OBS, Audacity, or a camera menu comes too late. The signal may already be boosted and clipped inside the transmitter. You are making a clipped signal quieter, not preventing the clipping in the first place.

The Hollyland LARK MAX 2 camera receiver gives you the option to control transmitters' gain directly from the unit. 

Additionally, you can use the HollyAudio app to adjust Dynamic Gain levels (High, Medium, or Low), or set custom gain from the Set gain to section for each transmitter.

As a practical target, aim for 6 to 10 dB of headroom below 0 dBFS during the loudest moments of a session. This is the core idea behind gain staging: setting each amplification stage so the signal is clean and has room to peak without clipping.

Step 2: Adjust Microphone Placement

If reducing gain does not fully resolve the problem, adjust the microphone position. Moving the mic 3 to 6 inches farther from the source, or angling it slightly off-axis from the speaker’s mouth, reduces the SPL reaching the capsule without losing intelligibility or audio quality. For lavalier and clip-on wireless mics, positioning the capsule slightly below the chin and a few inches off center is a reliable starting point.

Step 3: Lower Software Input Level

Only reach for the software input level control after addressing gain at the hardware stage. Reducing the input level in your recording application trims the signal going to the recorder, but it does not undo clipping that already occurred upstream in the chain. Also,  lowering software input raises the noise floor relative to the signal, degrading your signal-to-noise ratio. Use this control to fine-tune levels, not as the primary fix.

Step 4: Enable a Hardware Limiter or Pad

A limiter automatically catches signal spikes before they reach 0 dBFS, acting as an insurance layer for unpredictable loudness events. Many audio interfaces, field recorders, and wireless systems include an onboard limiter. Enable it as a safety measure rather than a substitute for proper gain structure.

A hardware pad (-10 dB or -20 dB) is a passive attenuation switch found on some microphones and interfaces. Use a pad when the input source is inherently loud, such as drums, guitar amplifiers, or a loud speaker at close range, and the standard gain control cannot be reduced far enough to bring the signal into a safe range.

How 32-Bit Float Recording Eliminates Peaking?

All of the fixes above are reactive. They work by preventing the signal from reaching 0 dBFS in the first place. But there is an architectural approach that changes the problem entirely: 32-bit float recording.

Standard digital audio uses 16-bit or 24-bit depth. These recording formats cannot capture audio beyond 0 dBFS. When levels go above that limit, the extra information is lost. The result is permanent clipping in the recording. Once that happens, editing tools cannot fully bring the missing audio back.

32-bit float works differently. It uses a floating-point number format that can represent a vastly wider dynamic range, far beyond what any microphone, voice, or acoustic instrument can physically produce. In practice, this means the converter captures the full analog signal even if the output level going to a camera or recorder would have clipped at 24-bit. There is no ceiling to hit. Any transient that would have peaked is stored accurately and can be pulled back in post-production without any clipping artifacts. Peaking becomes a non-destructive, recoverable event rather than a permanent loss.

The Hollyland LARK MAX 2 puts this into practice through its 32-bit Float Internal Recording feature. The LARK MAX 2 transmitters record audio locally at 48 kHz / 32-bit float as a backup while simultaneously sending audio to the connected camera or recorder. If the output to the camera peaks and clips, the 32-bit float backup file on the transmitter preserves the undistorted original signal in full.

Here is how that workflow looks in practice:

1. Record the session by pressing the REC button on the TX. Loud moments cause the camera’s audio track to peak and clip.

2. After the session, retrieve the 32-bit float backup file from the LARK MAX 2 transmitter. To do that:

• Turn off the transmitter by pressing and holding the multi-function button when the TX is powered on.

• Connect the TX unit to your PC via the provided USB-C cable. Each TX has its own dedicated USB-C port.

• Turn on the transmitter while the unit is connected to your PC. Your computer should automatically recognize it as an external drive.

• Open the drive, find the file, and copy and paste it into your PC. 

3. Import the backup file into your editing software.

4. Reduce the gain on the loud sections. Because no data was clipped at capture, the waveform is clean and pulls back without any distortion.

The LARK MAX 2 also includes OWS earphones for wireless monitoring that provide real-time, low-latency in-ear audio feed during recording. Using this lets you hear the live signal and catch a peaking level before committing to a full take, which works as a complementary prevention step alongside the 32-bit float feature.

Preventing Microphone Peaking: Gain Staging Best Practices

The best time to solve peaking is before it happens. These practices take only a few minutes before each session and eliminate most peaking problems entirely.

• Target peaks between -12 dBFS and -6 dBFS: For voice and dialogue, this range leaves enough headroom for unexpected loudness while keeping the signal clean and usable.

• Set your average (RMS) level around -18 dBFS: Peaks near -12 dBFS with an average near -18 dBFS is the standard gain structure for broadcast-quality spoken word recording.

• Always test at the actual loudest level: Have the speaker talk at their real maximum volume during the check, not a soft test count. Many peaks happen because the session started with a quiet test and the speaker got animated mid-record.

• Adjust the gain at the microphone first, then fine-tune the rest of the audio chain: Start at the transmitter or interface preamp, then verify the software input level. Never start at software and work backward up the chain.

• Use a limiter as a safety layer, not a crutch: A limiter should catch the occasional unexpected spike. If it is working constantly and compressing heavily, the gain is still set too high.

• Record a safety track if 32-bit float is not available: Many interfaces and recorders support dual-track recording: a main track at your target level and a safety track recorded 6 dB lower. This is a standard technique in broadcast and field production.

Can You Fix a Peaked Recording in Post?

The answer depends on how the audio was originally recorded.

For standard 16-bit or 24-bit recordings, hard clipping represents permanent data loss. The information above 0 dBFS was never stored, and no software can reconstruct what was never captured. Tools like iZotope RX and Adobe Audition can make clipped audio sound less harsh. They do this by rebuilding parts of the damaged waveform. But the repaired audio is only an estimate of what was lost. Results range from passable to still noticeably distorted depending on how severely and how long the signal clipped.

For 32-bit float recordings, the situation is fundamentally different. Because no data is lost at the capture stage, pulling back a peaked section in post simply means reducing the gain. The full waveform was preserved, so the result is clean and undistorted. This is why 32-bit float recording is not just a convenience feature. For anyone working with unpredictable audio levels, it changes peaking from a permanent problem into a minor editing step.

FAQs

Q: What is the difference between microphone peaking and audio clipping?

Peaking occurs when the audio level reaches the system's maximum limit. Clipping is the distortion you hear after that limit is exceeded. The waveform gets cut off, which damages the sound. Peaking always comes first. A very short spike may not cause obvious distortion, especially when a limiter is being used. But frequent or extended peaks will eventually produce clipped and damaged audio.

Q: What dBFS level should my microphone be recording at?

For voice and dialogue, target peak levels between -12 dBFS and -6 dBFS, with an average (RMS) level around -18 dBFS. This range keeps the signal strong and clean while leaving enough headroom to handle sudden loud moments without hitting the 0 dBFS ceiling.

Q: Why does my microphone keep peaking even after I lower the gain?

If reducing gain in your recording software does not stop the peaking, the problem is likely happening earlier in the signal chain. For wireless microphones, check the transmitter gain. If it is set too high, the signal is already clipping before it reaches the software. For wired setups, check the preamp gain on the interface. Work through each stage from source to software.

Q: Does a wireless microphone peak differently than a wired mic?

The basic behavior is the same for both systems. The difference comes from the extra gain stage inside wireless systems. Many users miss this part when setting levels. If transmitter gain is too high, lowering software input will not fix it. The signal gets clipped inside the transmitter before it reaches your recording app. Always adjust gain at the transmitter first.

Conclusion

Microphone peaking is a gain problem, and the solution always starts at the earliest point in the signal chain. Reduce hardware gain before adjusting software levels, and verify your placement. And treat a limiter as a safety layer rather than a level-setting tool. For wireless mic users, controlling the transmitter’s gain is the step most often skipped. If your setup supports 32-bit Float recording, you have a backup that makes peaked audio recoverable rather than lost.