Podcast Editing Workflow: From Raw Recording to Final Cut

Stage 1: Recording During session

Good audio starts at the recording stage. No amount of editing fixes fundamentally bad audio. The basics: close-microphone technique (3–6 inches from the mic), headphones for all speakers to avoid bleed, a quiet room, and recording each speaker on a separate track if possible.

For remote interviews, tools like Riverside.fm and Zencastr record local audio from each participant, avoiding the compression and artifacts of Zoom. This matters more than most people realize — Zoom audio that's been re-encoded sounds noticeably worse when edited, especially after noise reduction.

Key output: Raw WAV files for each speaker, recorded at 48kHz/24-bit or better. Keep backups immediately.

Stage 2: Intake and organization 15–30 min

Before you start editing, get organized. Create a project folder with a consistent structure:

• raw/ — original, unedited audio files (never touch these)
• edit/ — your DAW project files
• transcript/ — transcript files
• export/ — final rendered audio

Label your files consistently. "EP047-interview-jane-raw.wav" is much better than "recording (1).wav" six months later when you're looking for a clip.

Do a quick listen to check audio quality. Catch problems early: excessive background noise, technical failures, one speaker much louder than the other.

Stage 3: AI transcription 1–5 min

Upload your main audio file to a transcription service and get a word-level transcript. With modern AI transcription tools like Deepgram, this takes 1–2 minutes for an hour of audio. In Paper Edit, it's automatic on upload.

The transcript is the raw material for the next stage. Don't proofread or correct it yet — you're about to read through it as an editor, not a proofreader.

Stage 4: Structural edit (paper editing) 45–90 min

This is the heart of the workflow. Read through the transcript with editorial eyes: What's the story? What's the best material? What can go?

The paper editing process has four sub-stages:

1. First read: Get the lay of the land. No marking yet.
2. Markup: Highlight keepers, cross out cuts, flag maybes.
3. Script building: Arrange selected passages into narrative order.
4. Review: Read through the script as a whole. Does it hold together?

At the end of this stage you have a script — a precise list of audio segments, in order, with start and end times. This is the "paper edit."

Stage 5: Rough assembly 15–30 min

Convert the paper edit into an audio timeline. If you're using Paper Edit, this means exporting the Reaper RPP project file — all your clips are automatically placed on the timeline. If you're working manually, import your cuts into your DAW in order.

Do a quick listen-through of the assembled rough cut. Is the overall shape right? Are there gaps that need bridging narration? Make notes but don't make micro-edits yet.

Stage 6: Fine edit 30–60 min

Now you're in the DAW doing waveform work. This is where you:

• Tighten cut points — trim clips so they start and end cleanly
• Add crossfades between clips to smooth transitions
• Remove individual breaths, clicks, and mouth noises
• Fix any problems with the assembly: abrupt cuts, missing context, timing issues

This pass goes fast because all the structural decisions are already made. You're polishing, not deciding.

Stage 7: Music, sound design, and production 15–45 min

Add the production elements that define your show's identity: intro music, outro, transition music, ads/sponsorship reads, chapter markers if your feed supports them. This is where the episode starts to feel like a finished product rather than an edited conversation.

Keep music levels roughly 20–25 dB below speech. Music that competes with speech is one of the most common mistakes in podcast production.

Stage 8: Mix and master 20–40 min

The mix is about balancing everything — relative levels of all elements, EQ to remove problematic frequencies, dynamic control. The master is about the final loudness and loudness consistency. For podcasts:

• Target loudness: -16 LUFS integrated (Spotify, Apple Podcasts standard)
• True peak ceiling: -1 dBTP
• Noise floor: Below -60 dB is fine for most content

Tools like Auphonic can automate a lot of the loudness and noise reduction work. For a manually-mixed episode, a simple chain is: noise reduction → EQ → compression → limiter.

Stage 9: Export and publish 10–20 min

Render the final MP3 (128kbps stereo or 64kbps mono are standard). Add ID3 metadata: title, episode number, artwork, description. Upload to your podcast hosting platform and schedule. Done.

Keep your project files. You'll want them if you need to re-export, fix an error, or re-use a clip.