How to Reduce Background Noise in Podcast Recordings on Mac

How to Reduce Background Noise in Podcast Recordings on Mac

Background noise is the fastest way to make a good episode sound amateur. Fan hum, fridge drone, keyboard clicks and street rumble all get noticed, even when listeners can’t name the problem. The good news: you can reduce podcast background noise on Mac with free tools, as long as you match the fix to the type of noise and keep every step gentle.

Short answer: Record close to the mic with peaks around −12 to −6 dBFS and capture 10–20 seconds of silent room tone. In Audacity, use that room tone as the noise profile, run one Noise Reduction pass at about 6–12 dB, and preview the Residue to make sure none of your voice is being removed. Then add a high-pass filter, gentle compression and loudness normalization last.

Key takeaways

  • Identify the noise first: steady hiss, tonal hum, rumble and one-off sounds each need a different fix.
  • Clean capture beats any plugin, so set your gain, get close to the mic and soften the room.
  • Always record room tone. Audacity’s Noise Reduction can’t work without a clean noise-only sample.
  • Use the Residue preview: if you can hear your words in it, your settings are too strong.
  • Remove hum with a notch filter before Noise Reduction, and compress only afterwards.
  • Background voices, clipping and heavy echo can’t be fully fixed in editing.

First, Identify Which Kind of Noise You Have

Put on headphones and listen to a pause in your recording. What you hear there tells you which tool to reach for. Using the wrong one, such as a noise profile on a barking dog, is why many cleanups end up sounding worse than the original.

What you hear Typical cause Best first fix
Constant “shhh” or airy whoosh Laptop fan, air conditioning, preamp hiss Audacity Noise Reduction with a room-tone profile
Steady low pitched hum or buzz Mains hum (50 Hz in Europe, 60 Hz in North America) and its harmonics, ground loops Check cables and power first, then a notch filter, then Noise Reduction
Deep rumble or thumps Traffic, footsteps, desk bumps High-pass filter at 80–100 Hz
Sounds that come and go Keyboard clicks, doors, a dog barking Cut them out or re-record the line. A noise profile can’t learn them.
Ringing tail after each word Echo from bare walls Soften the room before the next session. Post-processing helps only a little.
Other people talking Housemates, café chatter Re-record if possible; voice-separation tools are a compromise

Why Podcast Noise Is Hard to Fix Later

Noise in the pauses is easy to hear and fairly easy to remove. Noise sitting under your voice is harder. Software can pull down a steady hiss because it can measure it on its own. It struggles when the noise changes over time, shares frequencies with speech, or is nearly as loud as your voice.

That’s why the order of operations matters:

  1. Record cleaner: gain, mic distance and room.
  2. Capture a noise profile: room tone with no speech.
  3. Remove hum, then reduce noise gently: one light pass beats several heavy ones.
  4. Filter, compress and set loudness last.

Skip the first step and no free or paid tool will fully rescue the take.

Step 1: Stop Background Noise Before You Hit Record

Every bit of noise you avoid at the source is noise you don’t have to remove later, so the cleanup can stay gentle and your voice keeps its natural tone.

Set your gain so peaks land around −12 to −6 dBFS

Watch the meters in your recording app and aim for speaking peaks of roughly −12 to −6 dBFS. Too quiet, and you’ll lift the noise floor when you boost later. Too hot, and you’ll get clipping, which no tool can undo.

Place the mic close and aim it away from noise

  • Get close: about a fist’s distance from the mic means more voice and less room, fan or traffic.
  • Use the null: a cardioid mic rejects sound from behind, so point its back toward the laptop fan or the window.
  • Move the Mac: place your MacBook beside or behind the mic so its fan sits outside the pickup zone.

Soften the room

Hard walls bounce sound around, and that echo is far harder to remove than hiss. Soft surfaces nearby, such as a sofa, curtains, a blanket or a full bookshelf, absorb reflections more than most people expect. Also switch off fans, close windows and silence notifications before you start.

Always record 10–20 seconds of room tone

At the start or end of every take, stay silent for 10–20 seconds with the same mic, position and gain you use for speech. This noise-only sample is what Audacity learns from in Step 2. Without it, you’ll be hunting for tiny gaps between sentences, and those usually contain breaths.

Use Voice Isolation for live calls

For calls in FaceTime and other supported apps, macOS offers a Voice Isolation mic mode. While the app is using your mic, open Control Center, click Mic Mode and choose Voice Isolation. According to Apple’s Mic Modes guide, the feature requires a Mac from 2018 or later, and the available options vary by app. It works in real time only, so it can’t clean a file you’ve already recorded.

Interviewing remotely? Ask each guest to record their own microphone locally while you talk, then line up the files in the edit. You get each voice without call compression or the other person’s room noise, and the call recording remains a backup.

Recording Mac system audio? Skip the second mic

If you capture system audio (a Zoom call, browser, game or reference track), don’t hold a phone up to the speakers. Route the internal audio directly so room noise never enters the recording. Our free walkthrough shows how with Easy Audio Recorder Lite and a virtual audio device: how to record internal audio on Mac for free.

Step 2: Free Podcast Noise Reduction on Mac with Audacity

Audacity is free, runs offline on Mac and handles steady noise such as hiss, fan noise and hum well. Its official manual states plainly that it isn’t meant for clicks or irregular noise like traffic, which is why the diagnosis table above matters. The effect works in two passes: it learns the noise from a noise-only selection, then lowers that noise wherever you apply it.

Optional first: remove mains hum with a notch filter

If you hear a steady pitched hum, fix the cause first: try another cable, outlet or USB port. If the hum remains, use Audacity’s Notch Filter (in the EQ and Filters group of the Effect menu) at 50 Hz in Europe or 60 Hz in North America, and on its stronger harmonics if needed. Do this before Noise Reduction.

Run Noise Reduction

  1. Duplicate your track (Edit → Duplicate) so you can compare against the original.
  2. Select your room tone: at least 2–3 seconds with no words and no breaths. A longer clean sample gives a more reliable profile.
  3. Open Effect → Noise Removal and Repair → Noise Reduction and click Get Noise Profile. Older versions list Noise Reduction directly in the Effect menu.
  4. Select the whole clip or just the noisy region.
  5. Reopen Noise Reduction, start from the settings in the table below and choose Residue under “Noise”. Click Preview.
  6. Check the Residue. You should hear only hiss and hum. If you can make out words or the shape of your voice, lower Noise Reduction or Sensitivity and preview again.
  7. Switch back to Reduce, preview once more, then click OK. Apply one pass only.
Setting Start at When to change it
Noise reduction (dB) 6 dB, up to about 12 dB Use the lowest value that makes the noise acceptable. Higher values damage the voice.
Sensitivity 6 (default) Lower it if your voice appears in the Residue; raise it slightly if “tinkly” artifacts remain.
Frequency smoothing (bands) 6 (default) Leave it as is for speech; higher values favor spoken word and can soften artifacts.

Good to know: Audacity forgets the noise profile when you quit, so capture a new one each session. If one recording has two different noises, for example a fan in the first half and an air conditioner in the second, profile and reduce each one separately.

Pro tip: Every few minutes, solo the processed track and then the duplicate. Your ears adjust to processing quickly, and a quick A/B comparison keeps you honest.

[AUTHOR INPUT NEEDED: Optional but valuable. Embed a 10-second before/after clip of your own recording with the exact settings you used.]

Step 3: Finish with a Light Cleanup Chain

Noise reduction alone rarely makes a voice sound finished. A short, gentle chain sounds more natural than one heavy effect. Apply it in this order:

  1. High-pass filter at 80–100 Hz (Effect → EQ and Filters → High-Pass Filter) removes rumble and desk thumps without thinning most voices.
  2. Light noise gate on long pauses only. Set it so it never clips word endings or quiet breaths inside phrases.
  3. Gentle compression at around 2:1 to 3:1 keeps quiet phrases audible without obvious pumping. Always compress after noise reduction, because compression lifts the quiet noise between words.
  4. Loudness normalization last. Apple Podcasts recommends about −16 LKFS (±1 dB) with true peaks no higher than −1 dBFS. Our step-by-step guide covers it: how to set podcast loudness on Mac.

In current Audacity versions, the gate, compressor and Loudness Normalization effects sit under Effect → Volume and Compression. New to these terms? Our audio editing basics for beginners explains cuts, levels and what “clean” really means.

Audacity vs. GarageBand vs. Paid Tools: Which Should You Use?

The right tool depends on the type of noise. Steady noise suits a classic noise profile. Changing noise often needs smarter processing or a new take.

Option Cost Best for Limits
Audacity Free Steady hum, hiss, fan and fridge noise via a noise profile Artifacts if pushed hard; not designed for irregular noise
GarageBand Free on Mac Noise gate, EQ and compression for simple cleanup No profile-based reduction for noise under the voice
Paid spectral or AI voice tools Varies Cafés, keyboards, variable or very loud noise Cost; some process audio in the cloud, so check privacy
Re-recording Your time Background voices, clipping, heavy echo Not possible for one-off interviews

Start with Audacity for steady noise. Move to a dedicated tool only when one mild pass can’t do the job, and ask yourself whether a second take would simply be faster.

What Noise Reduction Can’t Fix

  • Other people talking in the same room, because their voices share your frequencies.
  • Clipping and distortion from recording too hot.
  • Heavy reverb from a tiled bathroom or empty hallway. You can reduce it a little, but don’t expect a studio sound.
  • USB or cable buzz that changes when you move the mouse. Fix the hardware path first by trying another port, cable or a powered hub.
  • Noise almost as loud as your voice. The tool can’t separate the two cleanly, and heavy settings will damage the speech.

5 Common Podcast Noise Reduction Mistakes

  1. Stacking several heavy passes. Each pass adds artifacts; one mild pass almost always sounds better.
  2. Profiling a section with speech or breaths. The tool then treats parts of your voice as noise.
  3. Compressing before noise reduction. Compression lifts the quiet noise between words and makes clean removal harder.
  4. Skipping the Residue check. Without it, you only notice the damage after export.
  5. Editing on laptop speakers only. Small speakers hide hum and hiss, so always check with headphones.

Your Mac Podcast Noise Checklist

Run through this list before every episode:

  1. Peaks around −12 to −6 dBFS while speaking.
  2. 10–20 seconds of room tone captured.
  3. Hum fixed at the source or notch-filtered first.
  4. One mild Noise Reduction pass, checked with Residue.
  5. High-pass filter applied for rumble; gate working on pauses only.
  6. Compression after noise reduction; loudness set last.
  7. A/B check against the raw take on headphones.
  8. WAV exported for your archive, plus the format your host wants.

FAQ: Reducing Podcast Background Noise on Mac

What’s the best free way to reduce podcast background noise on Mac?

Record cleanly first, then use Audacity’s Noise Reduction with a room-tone profile, starting at about 6 dB and rarely above 12 dB. Preview the Residue to confirm your voice isn’t being removed. It’s free, works offline and handles steady hum and hiss predictably.

Should I remove noise before or after normalizing?

Simple normalizing or amplifying raises voice and noise equally, so it can come before or after. Compression and loudness processing should come after noise reduction, because they lift the quiet noise between words. The safest order is noise reduction, EQ, compression, then loudness.

Is GarageBand enough for podcast noise reduction?

For simple cases, yes. GarageBand’s noise gate can silence low-level noise in pauses. It has no profile-based noise reduction, though, so steady hum under your voice is usually cleaner with Audacity.

How do I remove electrical hum from a recording on Mac?

Check cables, outlets and USB ports first. If the hum remains, apply Audacity’s Notch Filter at 50 Hz (Europe) or 60 Hz (North America) and its stronger harmonics, then run Noise Reduction for whatever hiss is left.

Why does my voice sound robotic after noise reduction?

The reduction is too strong, Sensitivity is set too high, or the noise profile included speech or breaths. Lower the settings, pick a cleaner room-tone sample, and use the Residue preview to make sure no words are being removed.

Does macOS have built-in noise reduction?

macOS offers a Voice Isolation mic mode in Control Center for supported apps on Macs from 2018 and later. It helps on live calls, but it works in real time and can’t clean a file you’ve already recorded.

Clean Podcast Audio Starts Before the Edit

Reducing podcast background noise on Mac is mostly about habits and diagnosis. Set your gain, get close to the mic, record room tone, and match each fix to the noise you actually hear. Free tools like Audacity then handle steady hum and hiss surprisingly well. For everything else, a softer room or a quick re-record usually wins.

Ready to record cleaner audio? Grab Easy Audio Recorder Lite free on the Mac App Store for simple WAV recordings. When you want cutting, leveling and effects in one desktop workflow, take a look at the NeonWaves Mac audio editor.