Podcast Audio Prep for macOS
WaxOn/WaxOff is a two-mode audio tool for podcasters. WaxOn prepares raw recordings for editing: high-pass filtering, noise reduction, loudness normalization, phase rotation, and brick-wall limiting. WaxOff finalizes your edited mix for distribution: EBU R128 loudness normalization, true peak control, and MP3 encoding.
Download v1.3.1 (DMG) · Manual
⚠️ Important: Read Before First LaunchmacOS will block the app because it is not notarized with Apple. After dragging WaxOn to Applications, run this command in Terminal:
xattr -cr /Applications/WaxOnWaxOff.appWithout this step, macOS will refuse to open the app.
Use WaxOn on raw recordings before editing. Drop your files in, configure what you care about, and get to editing.
- High-Pass Filter — configurable cutoff (20–90 Hz, default 80 Hz) removes rumble, HVAC hum, and handling noise; set to 20 Hz for DC Block only
- Noise Reduction — optional RNNoise ML-based background noise suppression (off by default); best for recordings with consistent steady-state background noise. For stereo output, channels are split and denoised independently to ensure balanced processing
- Loudness Normalization — optional two-pass EBU R128 with configurable target (−35 to −16 LUFS); linear gain only, dynamics fully preserved. When NR is off, the analysis pass uses noise reduction internally to prevent broadband noise from skewing the measurement, keeping the output unmodified.
- Noise Floor Detection — estimated noise floor displayed in the stats panel with color-coded warnings; a
⚠️ badge appears on files with high noise floors that may affect loudness accuracy - Brick-Wall Limiting — 2× oversampled true peak control at the chosen ceiling (−1 to −3 dB)
- Phase Rotation — 200 Hz allpass to reduce peak asymmetry and improve limiter headroom
- Mono or Stereo Output — mono with left/right channel extraction, or stereo with per-channel noise reduction when NR is enabled
- Sample Rate Conversion — 44.1 kHz or 48 kHz output
- Mix — select 2+ files and combine them. When Loudness Norm is on, each file is individually leveled to the target LUFS before mixing, then the combined output is normalized to the same target, ensuring a balanced blend regardless of source levels
- Presets — five built-in presets (Defaults, Edit Prep, Edit Prep EBU, Mix 2 Channel, Mix Mono) plus custom presets saved and deleted from the toolbar menu
- Batch Processing — up to 3 concurrent jobs with per-file progress
Output: {name}-{rate}waxon-{limit}.wav (24-bit WAV)
Use WaxOff on your finished, edited mix. Apply broadcast-standard loudness normalization and deliver as WAV, MP3, or both.
- EBU R128 Loudness Normalization — two-pass analysis + linear gain; no dynamic processing, stereo image and transients are unchanged
- True Peak Control — configurable ceiling (−3.0 to −0.1 dBTP, default −1.0)
- WAV + MP3 Output — 24-bit WAV, CBR MP3 (128/160/192 kbps), or both; MP3 always outputs at 44.1 kHz
- Phase Rotation — optional 150 Hz allpass to reduce crest factor on bass-heavy material
- Presets — three built-in presets (Podcast Standard, Podcast Loud, WAV Only Mastering) plus custom presets
- Sample Rate Conversion — 44.1 kHz or 48 kHz
Output: {name}-lev-{target}LUFS.wav / .mp3
| Preset | Target | True Peak | Output | MP3 | Sample Rate |
|---|---|---|---|---|---|
| Podcast Standard | −18 LUFS | −1.0 dBTP | WAV + MP3 | 160 kbps | 44.1 kHz |
| Podcast Loud | −16 LUFS | −1.0 dBTP | WAV + MP3 | 160 kbps | 44.1 kHz |
| WAV Only (Mastering) | −18 LUFS | −1.0 dBTP | WAV only | — | 48 kHz |
- Waveform Preview — select a file to view its waveform with dB scale
- File Stats — format, sample rate, channels, bit depth, duration, bit rate, RMS, peak, crest factor, integrated LUFS, and estimated noise floor
- Noise Floor Warning — files with a high noise floor show a
⚠️ badge in the file list and a color-coded FLOOR stat (orange above −50 dBFS, red above −40 dBFS) - Drag & Drop — drop files anywhere on the window
- Resizable Layout — drag the divider between file list and waveform panel
- Custom Output Directory — optionally set a dedicated output folder
- Reveal in Finder — click to reveal processed files
- Delete Key — press Delete to remove selected files from the queue
- Independent File Lists — each mode keeps its own queue; switching modes doesn't disturb your work
- Collapsible Settings Sidebar — toggle the settings panel with the toolbar button; slides in/out from the right edge
Raw recordings → WaxOn → Edit in DAW → WaxOff → Distribute
WaxOn was designed for podcast audio, but the prep pipeline (high-pass filtering, phase rotation, loudness normalization, and limiting) maps well to any voice-forward production workflow. If you're editing interviews, documentary dialog, or other spoken-word content outside a full DAW environment, it works the same way.
WAV, AIFF, AIF, MP3, FLAC, M4A, OGG, Opus, CAF, WMA, AAC, MP4, MOV. FFmpeg is bundled; no separate installation required.
- macOS 14.0 (Sonoma) or later
- Apple Silicon or Intel Mac
Copyright © 2026. This app was designed and directed by Seven Morris, with code primarily generated through AI collaboration using OpenClaw and Claude (Anthropic).
This program is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License v3.0.
I'm a freelance audio engineer, not a software developer. These tools exist because AI made it possible for me to build things I couldn't build alone, and I think that's genuinely valuable.
But I hold that alongside some serious concerns. AI raises deep questions about labor displacement, resource consumption, surveillance, the concentration of power in a small number of corporations, and the increasingly close relationship between those corporations and governments. These aren't hypothetical risks; they're unfolding now, and the implications for ordinary people are significant. I don't have clean answers. I don't think anyone does.
What I can say is that I think it matters how these tools get used, and by whom, and toward what ends. A free audio utility that helps independent podcasters is one kind of use. There are others that are far less benign. I'd rather be honest about that tension than pretend it doesn't exist.
The current app icon is my own design. AI can build the software, but I can still make the art myself.
WaxOn/WaxOff has been built carefully and iteratively, tested in real podcast workflows, refined based on actual use, and updated continuously as improvements reveal themselves. That process takes real time and attention, even when AI is writing the code. The app is free and will stay that way. If you find it useful, a coffee is appreciated.
