Recording Guide
This guide covers advanced settings for getting the best possible audio from Rebel Audio's local recording pipeline.
Quality settings
Sample rate
Default: 48kHz. Recommended for all podcast use — this is the broadcast and streaming standard. Change to 44.1kHz only if your post-production chain requires it. Setting is per-session in Session Settings → Audio.
Bit depth
Default: 24-bit. Provides 144dB of theoretical dynamic range — more than enough headroom for voice recording. Use 16-bit only if disk space is a constraint (rare for podcast lengths).
Buffer size
Default: 4096 samples (at 48kHz = ~85ms). This is the IndexedDB write interval. Lower values reduce the risk of losing audio on crash; higher values reduce CPU overhead. Adjust in Session Settings → Advanced.
Gain staging
Target input level: -18 to -12 dBFS RMS. Leave headroom for peaks. Rebel Audio's auto-level-to--16 LUFS export applies post-gain correction, but starting with a well-staged signal reduces noise floor.
- Dynamic mics (Shure SM7B, Rode PodMic): set gain to 50–60% on interface
- Condenser mics: set gain lower, watch for clipping on sibilants
- USB mics: use system sound settings to set input level
Guest setup recommendations
Send your guest this checklist before the session:
- Use Chrome or Edge (latest version)
- Use headphones — prevents echo on their track
- Use a USB or built-in mic (no phone call routing)
- Close other browser tabs that use the microphone
- Check available disk space: a 90-minute session writes ~600MB per track
Browser permissions
Rebel Audio requires microphone access. Chrome will prompt on first session. If the prompt was dismissed:
- Click the lock icon in the address bar
- Set Microphone to Allow
- Reload the page
Troubleshooting local recording failures
If the waveform meter shows no movement after clicking Record:
- Check that the correct microphone is selected in Session Settings → Devices
- Open Chrome DevTools → Console. Look for
AudioContexterrors - Some antivirus software blocks IndexedDB writes — whitelist rebelaudio.org
- If the issue persists, use the Server backup mode toggle (Creator/Studio plans) — audio routes through server in compressed format as fallback
Track naming
Name each participant's track before recording starts in Session → Guest List. Track names carry through to exported file names: ep42-jared-host.wav, ep42-priya-guest.wav.
What Rebel Audio handles vs. your DAW
Rebel Audio captures, drift-corrects, and exports lossless stems. What it does not do:
- Dialogue compression / de-essing / EQ: handled in your post chain (Auphonic, Hindenburg, Logic, Adobe Audition, Reaper)
- Room-tone matching between hosts: requires per-track treatment in your DAW
- Final mastering: the -16 LUFS export is a first-pass target, not a finished master
- Noise reduction / restoration: if your guest's mic captures HVAC noise or keyboard clatter, that's in the WAV — we record the signal honestly
The export from Rebel Audio is the input to your post chain, not the output of it.