A podcast lives or dies on its audio. Listeners will put up with a slow intro or an off-topic tangent, but they stop listening when the sound is tiring: a hiss in one ear, one host much louder than the other, or a sudden jump in volume when the ad starts.
Here is a practical checklist we'd use on any episode, in the order that makes the most difference.
Before you record
The cheapest fix is the one you never need.
- Get the microphone close. A few centimetres from the mouth (with a pop filter) gives you far more voice relative to room noise than any setting can.
- Turn off what hums. Fans, air conditioning, fridges and noisy laptops. Silence notifications.
- Soften the room. Curtains, a sofa, a bookshelf, even a duvet behind the microphone cut echo, and echo is much harder to remove later than steady noise.
- Record each person on their own track if your setup allows it. Separate tracks let you fix one voice without touching the others.
- Leave headroom. Aim for speech peaks well below the maximum. A recording that is a little quiet is easy to fix; one that clipped is damaged for good.
- Record a few seconds of silence at the start. It tells you exactly what noise you are dealing with.
Cleaning the recording
Do the clean-up in this order, because each step depends on the one before.
- Remove background noise first. Hiss, fans, air conditioning and street noise make every later step harder. Our speech AI keeps the voice and removes the sound around it, so this is one step rather than an hour of tweaking.
- Check for echo and hum. Noise removal does not remove room reverb; if a guest recorded in a bare room, that is something to fix at the source next time. Electrical hum or buzz is best fixed at the source too, or with a hum-removal filter in your editor.
- Cut the obvious problems. Long silences, false starts, the cough you don't want. Edit on the cleaned audio so you hear what listeners will hear.
- Even out the voices. If one host is much quieter, bring them up so the listener isn't riding the volume.
- Set the final loudness. Spoken-word podcasts are commonly delivered around −16 LUFS (integrated), with true peaks kept below −1 dBTP so the file doesn't distort after the platforms re-encode it. Our tool can do this for you with the Podcast loudness preset.
Before you publish
- Listen to the start, a middle section and the end on headphones and on a phone speaker. Those are the two ways most people will hear it.
- Check the transitions. Intro music, ad breaks and the outro should sit at the same perceived loudness as the conversation.
- Export in a sensible format. MP3 is still the safest choice for podcast feeds. If you want a master to keep, save a WAV as well.
- Name your files consistently so the next episode is quick to find and compare.
A quick reference
| Problem | Fix |
|---|---|
| Hiss, fan, AC, traffic | Noise removal |
| Electrical hum or buzz | Fix the cabling or power; a hum-removal filter in your editor |
| Echoey room | Treat the room or move the mic closer next time |
| One voice louder than another | Separate tracks and level them |
| Episode too quiet or too loud | Loudness target (around −16 LUFS for speech) |
| Distorted, crackly peaks | Record with more headroom next time |
Do it in one pass
If you just want the noise gone and the loudness right, drop the episode into the podcast audio cleaner, pick the Podcast preset, preview a section and download. Video podcasts work the same way: the picture is copied as-is and only the audio is replaced.
Good audio isn't about expensive gear. It is about removing the things that make listening tiring, and this checklist covers most of them.