
In short: Edit a podcast in four passes, in this order. Clean the recordings first, then tighten the talk, then mix voices and music, then master the finished file. Doing it in that order stops each step from making the next one harder, and it gives you a routine you can repeat every episode.
Edit a podcast in four passes: clean, tighten, mix, master. Each pass has one job. If you do them in this order, you never undo your own work.
Most first-time editors do everything at once. They cut a word, add compression, lift a level, add music and then discover a hum under the whole recording. Fixing the hum now means redoing the compression and the levels. A fixed order avoids that.
Pass 0: prepare the session
Before the first pass, spend ten minutes getting organised. It saves more time than any plug-in.
- Back up the raw files and never edit the originals. Work on copies.
- Name the tracks by person, so "Host", "Guest 1" and "Guest 2" are obvious.
- Line the tracks up. If you recorded multitrack, the files should start together. If you recorded remotely, check the sync clap or the start points. Our podcast test pack includes a sync clap you can use to check alignment.
- Listen through once, start to finish, at normal speed, with a notepad. Note anything you want to fix, such as a loud cough at 14 minutes or a phone buzz near the end.
Pass 1: clean
Clean means removing faults from each recording before you touch the content. Do it track by track, not on the mixed result.
- Remove noise. Steady background noise such as a fan, air conditioning or a computer can be reduced with a noise reduction tool. Use it lightly. Heavy settings make voices sound watery.
- Fix hum and rumble. A high-pass filter set low, just above the rumble, removes low-frequency thumps from desks and doors without changing the voice.
- Deal with bleed. If one microphone picks up another person, check the tracks separately. Muting the quiet track between its speaker's words reduces the echo effect.
- Repair clicks, pops and plosives. Zoom in and fix the worst ones. The best cure is good mic technique at the recording stage.
- Even out sudden jumps. If one person leans in and gets much louder, adjust the clip level by hand rather than reaching for a compressor.
Clean first because every later tool makes noise louder. Compression lifts quiet sounds, including the fan. Normalising raises the whole file, including the hum. Remove the problem and the later steps behave.
Pass 2: tighten
Tightening is where you edit the content. You are shaping the conversation so it holds the listener, and you still want it to sound like people talking.
- Cut the long silences. Leave a pause so speech breathes, and take out the dead air. On the Generationally Speaking podcast we make, any silence over 1.6 seconds is cut down to 0.3 seconds, which keeps the pace without clipping a thought. Pick your own numbers and apply them the same way every time.
- Remove false starts and repeated words, where a speaker restarts a sentence or says the same word twice.
- Take out the distracting fillers and leave the natural ones. Cutting every "um" leaves audible joins and makes people sound like a machine.
- Cut the tangents and retakes. If the host asked the question twice, keep the better one. If a story wandered for two minutes without a point, trim it.
- Check each cut. Listen to every join. A cut inside a breath or a rising sentence sounds wrong, so move it to the end of a phrase.
Work from the biggest cuts to the smallest. Big structural cuts change the length and shape of the episode, so do those before you spend time on tiny fixes inside sections that may disappear.
Cut on the quiet part
When you cut inside a word or on a breath, you hear a click or a jump. Cut in the gap between phrases and add a very short fade to each side of the join. Most editors do this automatically, and it is worth checking.
Pass 3: mix
Mixing balances the sounds against each other. You are not changing the content now, only how it sounds.
- Set the voice levels. The host and guest should sound equally loud and close. Move the clip levels first, then add gentle compression to smooth the rest.
- Shape the tone. A small cut in the low-mids can reduce boxiness, and a small lift in the top end adds clarity. Make small moves, since a heavy EQ shows up quickly on headphones.
- Add music and effects. Place the intro, outro and any stings. Lower the music under speech and bring it back up in the gaps. Many editors do this with volume automation by hand, since an automatic ducking tool can pump.
- Check the transitions. Make sure the music finishes cleanly and that no sound cuts off abruptly.
Compare against a podcast you admire as you mix. Your ears will drift after an hour, and a quick reference listen resets them.
Pass 4: master
Mastering is the last stage and the only one that treats the finished episode as a single file. It sets the overall loudness and prepares the export.
- Set the loudness to the target for your platform. Our page on loudness targets for podcasts and platforms gives the numbers.
- Leave headroom so the loudest peaks do not distort when the file is converted to a compressed format.
- Export in the right format and bitrate. The file size and bitrate calculator shows what each choice costs in megabytes.
- Tag the file with the title, show name, episode number and artwork before you upload.
Master last because every earlier change alters the loudness. Set it once at the end and it stays set.
Check it before you publish
Listen to the finished file once on something other than your edit headphones: phone speaker, earbuds, car or laptop. Problems that hide on good headphones show up there. Listen to the first two minutes and the last two minutes with fresh ears, since those are where most listeners decide to stay or leave.
Common mistakes
- Mastering too early. Loudness changes with every edit, so do it last.
- Heavy noise reduction. It leaves a processed sound that tires listeners.
- Cutting every pause. Speech needs gaps to be understood.
- Mixing on one pair of speakers. Check on at least two.
- Editing the only copy. Always keep the raw files.
If you would rather record in a room built for it, with each voice on its own track, see the Podcast Studio. Separate tracks and a quiet room make pass one short.
Related guides
- Loudness targets for podcasts and platforms
- File size and bitrate calculator for podcasts
- All guides and tools
The picture at the top of this page is an illustration, not a photograph of a client or a real job.
Questions we get asked
What order should I edit a podcast in?
Clean first, then tighten, then mix, then master. Cleaning removes noise and faults before anything amplifies them, tightening cuts the content, mixing balances voices and music, and mastering sets the final loudness and export.
Should I remove um and ah from a podcast?
Take out the ones that slow the listener down, such as a long hesitation or a repeated word, and leave the natural ones. Cutting every filler makes speech sound robotic and leaves audible joins.
How long should the silences be in a podcast edit?
Short enough to keep the pace and long enough to sound human. On our own Generationally Speaking podcast, silences longer than 1.6 seconds are shortened to 0.3 seconds, which keeps the conversation moving without clipping anyone's thought.
granary.digital · 01243 884993 · studio@granary.digital · The Old Grainstore, Adsdean, Chichester, West Sussex, PO18 9DN · Prices exclude VAT unless marked inc VAT
