A spoken-voice chain has a different goal from a music vocal chain. No beat to cut through, no pitch correction, no creative effects: just a voice that's clear, even and comfortable to listen to for an hour. That makes it simpler, but consistency matters even more.
- Clean up noise first, then high-pass around 80 Hz
- Cut boxiness, lift presence gently
- Compression 3:1, around 4–6 dB on louder words
- De-ess, then a limiter to catch peaks
- No reverb; match a loudness target like −16 LUFS
The podcast vocal chain
- Noise reduction for constant hiss or hum. Gentle settings; heavy noise reduction sounds watery.
- High-pass filter around 70–100 Hz to remove rumble and desk bumps.
- Gate or expander (optional) to lower room noise between sentences. Keep it gentle so word endings aren't cut.
- EQ: cut boxiness around 300–500 Hz, add a little clarity around 3–5 kHz. Frequency guide: vocal EQ cheat sheet.
- Compressor: about 3:1, medium attack and release, 4–6 dB of gain reduction on louder words.
- De-esser for sharp "s" sounds: how to de-ess vocals.
- Limiter last, to catch peaks so you can raise the overall level safely.
Loudness targets
Podcasts are usually delivered around −16 LUFS integrated loudness (stereo), with true peaks no higher than about −1 dBTP. That's Apple Podcasts' recommendation and a common standard elsewhere. Platforms like YouTube normalise playback loudness, so hitting a sensible target avoids your episode being turned down or sounding quiet next to others. Measure with a loudness meter on the master.
Recording tips that matter more than plugins
- Record in the deadest room you have; reverb can't be removed later. More in how to record vocals at home.
- Keep a consistent distance from the mic. Dynamic mics are forgiving in untreated rooms.
- Each speaker on their own track and their own chain.
- Peaks around −12 to −6 dBFS while recording, never clipping.
Can you use music vocal presets on a podcast?
Usually not as-is. Music chains (including ours) include pitch correction, reverb, delay and character effects that a spoken voice doesn't want. If you use one, bypass the tuner and the time-based effects and keep the EQ, compression and de-esser stages. For voice-only content, a simple chain like the one above is the better starting point.
FAQ
What's the best compressor setting for podcast voice?
Around 3:1 with medium attack and release, so louder words come down 4–6 dB. Follow it with a limiter for peaks.
Should podcast voice have reverb?
No. Listeners expect a close, dry voice. Reverb makes it sound distant.
How loud should a podcast be?
About −16 LUFS integrated for stereo, with true peaks under −1 dBTP, is the common target.
