Capstone: Cleaning Up and Mixing a Podcast Episode

Every prior chapter covered one piece of a standalone audio workflow. This capstone takes a raw, two-host podcast recording all the way through that workflow — reading the waveform, cleaning it up, shaping its tone, evening out its dynamics, mixing in intro music, and exporting a finished episode — touching every chapter's own technique in the order a real editing session would actually use them.

Step 1 — Reading the Raw Waveform (Chapter 2)

The raw recording is opened and read visually first: one host's track runs noticeably quieter than the other's, a few flat-topped peaks show where a laugh briefly clipped, and several thin, near-flat stretches mark long pauses and a stretch of room tone recorded before either host started speaking.

Step 2 — Trimming Dead Air (Chapter 2)

The longest pauses and a false start at the very beginning are selected directly on the waveform and deleted, closing the gaps automatically — no ripple/roll trim-type decisions needed, since there's no video picture underneath to keep in sync.

Step 3 — Noise Reduction (Chapter 3)

The stretch of room tone identified in Step 1 is selected and sampled as a noise profile, then applied across the full recording — removing the room's own faint background hum before anything else happens to the file.

Step 4 — Normalizing (Chapter 3)

With the noise already removed, both hosts' tracks are normalized to a consistent target level — done after cleanup, not before, so the target volume reflects the actual wanted signal rather than a mix of signal and noise.

Step 5 — EQ: Clarity and De-Essing (Chapter 4)

A small cut in the low-mid range clears up a slightly boxy quality on one host's microphone, and a narrow cut around the sibilance band tames the other host's harsh "s" sounds — two targeted fixes, not a single blanket adjustment applied to both tracks identically.

Step 6 — Compression and Makeup Gain (Chapter 5)

This is where the loudness gap between the two hosts, spotted back in Step 1, actually gets addressed: a moderate threshold and ratio narrow that gap so neither host constantly needs the listener to reach for the volume knob, followed by makeup gain to bring the now-narrower dynamic range back up to a comfortable overall level.

Step 7 — Layering and Ducking Intro Music (Chapter 6)

A short intro music track is added on its own track, sitting several decibels below the hosts' own dialogue, with a simple ducking pass lowering it further under the opening spoken introduction and letting it fade out once the conversation begins in earnest.

Step 8 — Exporting for the Podcast Host (Chapter 7)

The finished mix is exported as an MP3 at 160 kbps — comfortably within the sensible range for spoken-word content — ready for direct upload to a podcast hosting platform, rather than as a WAV, since this episode's own final destination is a listener's feed, not further editing in another project.

Hands-On Exercises