No. Audio Polish only sets loudness and controls peaks (plus gentle clean-up for speech). It doesn't EQ, compress or change the mix the way mastering does.

What does voice levelling do?

It brings quiet and loud stretches of speech toward the episode's average by up to 6 dB, smoothly over about 1.5 seconds. It measures loudness, not who is speaking.

Why the 10-minute limit?

The whole file is held in your browser's memory while it's processed. Ten minutes of stereo audio keeps that comfortable on phones as well as laptops.

Is my audio uploaded?

No. Everything happens in your browser and the WAV is saved straight from it.

Audio Polish

Fix the level of an episode, a song or a clip in your browser: pick a preset, and get back a WAV at the right loudness with its peaks under control. Up to 10 minutes; the audio never leaves your device.

Rumble filter (80 Hz high-pass), gentle noise reduction between phrases, voice levelling (up to ±6 dB), loudness to -16 LUFS (Apple Podcasts' figure), true peak at or under -1 dBTP.

Measured entirely in your browser — your audio is never uploaded anywhere.

Your audio is decoded, processed and written to a WAV entirely in your browser. It is never uploaded or stored by CAPTD and never leaves your device.

Do this automatically from your episode audio

Spotlight's Episode mode uses the same clean-audio chain for its clips and audiograms, and writes the posts for every platform from the episode.

Episode mode is part of the Spotlight workspace on CAPTD's paid plans — see pricing.

Open Episode mode

How to use this

What it does

Decodes the file in your browser and runs one of three presets: Podcast voice (rumble filter, gentle noise reduction, voice levelling, -16 LUFS, -1 dBTP), Music loudness match (-14 LUFS and a peak limit only) or Social clip (rumble filter, -14 LUFS, -1 dBTP). Loudness is measured per ITU-R BS.1770-4. The result is a PCM WAV, 16-bit with dither or 24-bit.

What you'll need

An audio file up to 10 minutes: WAV, FLAC, AIFF, MP3, M4A, Ogg.
The preset that fits it.

Steps

Pick a preset.
Drop the file on the box.
Choose 16- or 24-bit and press Polish.
Listen to the result, then download the WAV.

Reading the result

Before and after show the loudness and true peak the preset started from and ended at. The steps list says exactly what was done to the file.

Tips

Use Music loudness match to compare mixes at the same level, or to hand a quiet bounce to a client — it doesn't change the mix itself.
For a long episode, polish it in sections up to 10 minutes, or export a WAV and use Spotlight.
Keep your original: polishing writes a new file.

Learn more

What does Audio Polish do — and not do?

It sets the level: loudness to a target in LUFS, peaks held under -1 dBTP with a look-ahead limiter, and, for speech, a rumble filter, light noise reduction between phrases and slow voice levelling. It's the same chain Spotlight uses for clips, with levelling added.

It doesn't shape the sound. There's no EQ, no multiband compression and no stereo processing, and the Music preset changes nothing but the level and the loudest peaks. It isn't mastering — a mastering engineer makes decisions this tool doesn't.

Why creators use it

Most level problems are simple: an episode far quieter than the platform norm, a clip that's too hot, a co-host much quieter than the host. Fixing them shouldn't need a plugin chain.

Common mistakes

Polishing a clipped file. levelling can't undo clipping; fix it at the source.
Using the podcast preset on music. noise reduction and levelling are for speech; music gets loudness and peaks only.
Polishing twice. run it once on the original, not on an already polished file.

FAQ

Is this mastering?: No. Audio Polish only sets loudness and controls peaks (plus gentle clean-up for speech). It doesn't EQ, compress or change the mix the way mastering does.
What does voice levelling do?: It brings quiet and loud stretches of speech toward the episode's average by up to 6 dB, smoothly over about 1.5 seconds. It measures loudness, not who is speaking.
Why the 10-minute limit?: The whole file is held in your browser's memory while it's processed. Ten minutes of stereo audio keeps that comfortable on phones as well as laptops.
Is my audio uploaded?: No. Everything happens in your browser and the WAV is saved straight from it.

Sources — checked on 20 October 2026

The Social clip target (-14 LUFS) is CAPTD's choice; no social platform publishes a loudness target.