How to Clean Up Podcast Audio for Free

Most podcast episodes do not need a studio to sound professional. They need four problems fixed, in the right order: background noise, filler words, awkward pauses, and uneven levels. Every one of these can be handled with free tools — you do not need a subscription, an engineer, or a treated room. This guide walks through the order that actually works, using UmBreak for the heavy lifting and free desktop editors for the finishing pass.

The right order (it matters)

Cleanup steps build on each other, so do them in this sequence:

1. Remove background noise first. Hum, room tone, hiss and fan noise make transcription less accurate, so clean them before anything else. UmBreak’s noise remover handles the common cases in the browser. If the noise is extreme, the free desktop app Audacity’s noise-reduction effect is the classic fallback — sample a few seconds of pure noise, then apply.
2. Remove filler words. Drop the cleaned file into the filler word remover. On-device AI transcribes the episode with a timestamp on every word and highlights um, uh, like, you know and similar fillers. Remove them in bulk, then click back any fillers that sound natural — zero-fillers speech sounds robotic.
3. Trim awkward pauses. Dead air between sentences is where listeners check out. UmBreak flags silences so you can tighten them (the full method is in the awkward-pause guide). Keep short natural breaths — cut only the gaps that serve no rhythm.
4. Even out the levels. If one host is louder than the other, run a normalisation pass in a free editor: Audacity’s “Normalize” and “Compressor” (in that order) get you 90% of the way to a consistent loudness. This is the step most beginners skip, and it is the one listeners notice most on headphones.
5. Export and check. Export as MP3 (128–192 kbps is plenty for speech) and listen to the first two minutes on earbuds and a phone speaker. Most amateur-sounding episodes fail on one of those two.

A forty-minute episode usually takes under twenty minutes this way.

What free tools cannot fix

Three problems no cleanup tool repairs well. Severe room echo — the “recorded in a bathroom” sound — is the big one; noise removers reduce it but cannot remove it. Fix it at the source by recording closer to the microphone in a smaller, furnished room. Heavy crosstalk (two people talking over each other) confuses filler and silence detection, so leave those sections alone or re-cut them by hand. Distorted, clipped audio where the waveform is crushed flat cannot be un-distorted; re-record with the input level lower.

The honest takeaway: free tools turn a decent recording into a good one. They cannot turn a bad recording into a decent one. Record clean, then clean up.

Record so future episodes need less cleanup

Five minutes of recording discipline saves an hour of editing:

• Mic distance: 10–15 cm from your mouth, slightly off-axis to avoid plosives.
• Kill the noise before it is recorded: turn off fans, close windows, record away from hard walls if you can.
• Record a few seconds of silence at the start — it gives noise-reduction tools a clean sample.
• Pause instead of filling: a silent two seconds is easy to trim; a two-second “uuuuuh” is not. See how to stop saying um.
• Record guests separately where possible; it makes every later step cleaner.

Cheat sheet: the full free workflow

Noise remover → filler word remover → silence trim → normalise + compress → export MP3 → earbud check. That is the whole process, free, with no account anywhere in the chain. For a deep dive on any single step, the guides are linked above — each one includes the honest limits of its method.