Podcaster's audio workflow — record, master, export to MP3, upload

The short version

1. Record to WAV or AIFF at 48 kHz / 24-bit. Don't record to MP3 — you'll compress twice and the result is mushy.
2. Edit in Audacity, Reaper, Audition, or Descript. Cut, trim, remove filler words.
3. Master to -16 LUFS integrated loudness for mono podcasts, -19 LUFS for stereo. True-peak at -1 dBTP max.
4. Export master WAV from your DAW at 48 kHz / 16-bit or 24-bit.
5. Convert to MP3: 96 kbps mono for pure voice, 128 kbps stereo if you have music.
6. Embed ID3 tags (title, artist, album, episode number, cover art 3000x3000 JPG).
7. Upload to host (Buzzsprout, Transistor, Spotify for Podcasters, Simplecast, Libsyn).

That's the seven-step chain. Our tooling runs step 5 — the WAV-to-MP3 conversion — entirely in your browser. Everything else stays with whatever DAW you prefer.

Step 1 — Record, not re-record

Record to uncompressed WAV (or AIFF on Mac, same thing different header). 48 kHz / 24-bit is the safe default. 44.1 kHz is also fine — it's CD quality and matches every distribution endpoint. What you want to avoid is recording to MP3 at 128 kbps, editing, then re-exporting to MP3 at 128 kbps. Every MP3 pass drops quality and voice artifacts stack up.

Remote interviews complicate this. Zoom, Riverside, and Squadcast record separately. Zoom's local M4A is OK but lossy; Riverside and Squadcast both record lossless WAV per-track locally. Use Riverside or Squadcast if quality matters; use Zoom if schedule matters.

Step 2 — Edit (we don't do this)

Editing is subtractive and creative, not mechanical. Honest tool picks:

• Audacity (free). Reliable, reference-open-source editor. Spectral view is excellent for finding clicks and mic bumps.
• Reaper ($60 personal / $225 commercial). The professional budget pick. Cheap, fast, scriptable. Most independent podcasters migrate here after Audacity.
• Adobe Audition ($20.99/mo). Industry standard for broadcasters and radio-adjacent podcasts. Multi-track session, spectral repair, loudness meter built in.
• Descript ($15/mo). Edit audio by editing a text transcript. Massive time-saver for dialog-heavy shows. Overdub voice clone is useful for fixing flubs.
• Logic Pro ($199 one-time, Mac). Full DAW; overkill unless you also compose music.

Step 3 — Master to -16 LUFS (the number that matters)

"LUFS" stands for Loudness Units relative to Full Scale. It's a measurement of perceived loudness averaged over the whole episode, and it's the number every podcast platform normalizes to on playback. Podcast standards:

• -16 LUFS integrated, mono. The Apple Podcasts spec, de-facto standard for single-speaker and voice-only shows.
• -19 LUFS integrated, stereo. Slightly quieter because stereo perceived loudness is ~3 dB higher than mono at equal LUFS.
• -1 dBTP true-peak ceiling. Never clip. True-peak measures the inter-sample peaks that can appear after lossy encode.

Hit those two numbers and your episode sits at the same perceived volume as NPR, Joe Rogan, and every other podcast in the queue. Hit them wrong and listeners crank the volume for your show and then get blasted by the next one. Every DAW has a loudness meter — Audacity needs the free Loudness Normalization effect; Reaper, Audition, Logic, and Pro Tools ship with one.

Tip: apply compression gently (4:1 ratio, ~3 dB reduction) before normalizing, not after. Compression evens the peaks so the normalization isn't fighting one loud laugh.

Step 4 — Export master WAV

Export the edited, mastered session as WAV at 48 kHz / 16-bit or 24-bit. Stereo vs mono depends on source:

• One-mic voice podcast: export mono. Saves 50% file size at every downstream step.
• Two mics panned hard L/R: still export as mono. Stereo separation of two voices sounds weird on headphones. Fold to mono in your DAW.
• Voice plus music stings or ambient bed: export stereo. Music benefits from the stereo image; you need stereo-capable encoding for those passages.
• Field recording, binaural, or ASMR: keep stereo.

The default is mono. 95% of podcasts are people talking into a mic, and mono is the correct answer for talking into a mic.

Step 5 — Convert to MP3 (the high-leverage step)

This is where our browser-side WAV to MP3 tool replaces an ffmpeg install or an Audacity export pass. Drop the WAV, pick a bitrate, download the MP3. Under the hood we run LAME at high quality with joint-stereo off (for mono) or on (for stereo). Real numbers:

The lazy Google answer of "192 kbps stereo" is exactly twice the bytes a voice podcast actually needs. Over 10,000 downloads of a 60-minute episode that's 290 GB instead of 432 GB — real money on host bandwidth tiers. Libsyn, Buzzsprout, and Transistor all meter bandwidth.

For the deep dive on encoder settings and LAME quality flags, see our WAV to MP3 guide.

Mono vs stereo — the part every tutorial gets wrong

If you have one person talking into one mic, record mono, export mono, encode MP3 mono. Stereo two-channel of a mono source doubles your file size and the encoder wastes bits encoding silence in the other channel.

If you have two people with two mics in the same room, still mix down to mono before encoding. Panning host L and guest R sounds disorienting on headphones. Pan both to center.

Go stereo only when there's a real stereo asset — a music intro with L/R instruments, a binaural field recording, a live concert. Those are rare in podcast-land.

Step 6 — ID3 tags and cover art

Every MP3 should ship with:

• Title: episode title.
• Artist: podcast name (NOT the host's name — use the show name).
• Album: season name or podcast name if you don't have seasons.
• Track number: episode number.
• Year: publish year.
• Genre: "Podcast".
• Cover art: 3000x3000 JPG, under 500 KB, embedded as APIC frame.

Most podcast hosts will overwrite or supplement these with their own metadata from your feed, but embed them anyway — listeners who sideload a file into a car head unit or offline player see the ID3 tags directly.

Step 7 — Upload to host

Buzzsprout ($12/mo up to 3 hours), Transistor ($19/mo unlimited), Simplecast ($15/mo), Libsyn ($5/mo tiny tier), Spotify for Podcasters (free, with caveats), Podbean, RSS.com. Pick one, upload the MP3, paste the show notes, hit publish. The host generates the RSS feed that Apple Podcasts, Spotify, Overcast, Pocket Casts, Castro, and the rest of the ecosystem pull from.

A real 60-minute voice podcast, timed

1. Record (60 min, live): WAV at 48/24 mono.
2. Edit in Reaper (45 min): remove mistakes, umms, dog barks.
3. Master (10 min): compressor, loudness normalize to -16 LUFS.
4. Export WAV (20 sec): 48 kHz mono WAV.
5. Convert to MP3 in browser (15 sec): 96 kbps mono.
6. Tag + cover art (2 min): drag cover, fill title.
7. Upload to Buzzsprout (90 sec): 43 MB MP3 goes fast.

Hands-on finishing time after recording: under an hour. The MP3 conversion step is where most tutorials lose 5-10 minutes to fighting Audacity's export dialog or installing ffmpeg. Browser tab + drag = 15 seconds.

Remote interview? Extract the audio from video

Zoom, Riverside, Squadcast, and SquadCast sometimes record to MP4 video even when you only want the audio. Don't burn an encoder pass — extract the audio track directly. Our audio extractor does a stream-copy when possible (AAC-in-MP4 comes out lossless and near-instant) and falls back to re-encode only if the target format differs. See the extract vs convert breakdown.

How our pipeline compares — honestly

Honest summary: use Audacity or Reaper for the edit, Auphonic for the automatic loudness + metadata pass if you want to pay for the shortcut, our WAV to MP3 converter for the final encode if you're manual. Don't use Online-Audio-Converter — it uploads an hour of your voice to a stranger's server.

Works well / doesn't work

Works well

• WAV/FLAC/AIFF master to MP3 at podcast-appropriate bitrates
• Preserving existing ID3 tags through conversion
• Extracting audio from Zoom/Riverside MP4 recordings (stream-copy lossless)
• Privacy-sensitive interview recordings (no upload)
• Batch-converting whole seasons in one folder drop
• Podcast-specific bitrate presets (96 kbps mono, 128 kbps stereo)

Doesn't work (well) yet

• Loudness normalization to -16 LUFS — on the roadmap; currently use your DAW
• Automatic filler-word removal — Descript and Auphonic own that job
• Noise reduction / denoise — use iZotope RX, Adobe Enhance Speech, or Krisp
• Multi-track editing — we're a converter, not a DAW

Common questions

What bitrate for a voice-only podcast?

96 kbps mono. That's the sweet spot where you hit transparent voice quality and stop paying for bits the encoder wastes on silence or noise floor. Don't go higher unless you have music; 128 kbps stereo is appropriate if you do.

Mono or stereo for two hosts?

Mono. Panning host L and guest R sounds weird on headphones. Mix to mono in your DAW, export mono, encode mono. Your file is half the size and listeners don't get disoriented.

What's -16 LUFS and why do I care?

It's the target loudness Apple Podcasts and Spotify normalize to. Master to -16 LUFS integrated and your show sits at the same perceived volume as every other podcast in the listener's queue. Ignore it and listeners yank the volume up for your show then get blasted by the next one.

Should I record in MP3 to save space?

No. Record to WAV. You'll compress to MP3 once at export, not twice. Two MP3 passes stack artifacts — sibilants get harsh, low end gets mud.

What sample rate — 44.1 or 48 kHz?

48 kHz. Matches video sync standards and most modern DAW defaults. 44.1 kHz (CD standard) is fine if your DAW is set to it. Don't switch mid-project.

Do I need a lossless master?

Yes, archive the edited WAV. Disks are cheap, re-editing from a lossy MP3 is not. A 60-minute voice-only WAV is ~330 MB mono or 660 MB stereo at 48/16.

My Zoom recording is MP4 — how do I get the audio out?

Use our audio extractor. It stream-copies the AAC track from MP4 without re-encoding when possible. That's lossless and fast. See the extract-vs-convert breakdown for when to re-encode.

Ready?

Record in your DAW, edit, master to -16 LUFS, export WAV. Then drop it into our WAV to MP3 converter, pick 96 kbps mono or 128 kbps stereo, download the MP3, tag it, upload to your host. Free, in your browser, no upload, no watermark. Remote interview came out as MP4? Start at extract audio from video instead — it'll stream-copy lossless in seconds.