Local-first recording. No network, no compromise.
Each track writes directly to your guest's disk in uncompressed WAV. Connection drops don't delete takes.
How local recording works
WebRTC carries the real-time video preview. The audio pipeline is separate — a Web Audio API node writes PCM samples to an IndexedDB buffer, flushed to a Blob every 30 seconds. At session end, the browser packages it as a BWF/WAV file and uploads it to the shared project. Your guest never notices the difference. You get 48kHz / 24-bit stems.
Drift correction: how it works
Oscillator drift between two machines averages 5–15ms per hour under normal load. Rebel Audio's alignment algorithm compares timestamps embedded in the audio stream against a shared session clock, applies a time-stretching correction pass (< 0.01% pitch shift), and re-renders before download. For a 90-minute interview the typical correction is 80–120ms — invisible in the waveform, audible if you skip it.
Export exactly what your post-production expects
Drop the stems straight into Auphonic for loudness matching, Hindenburg for assembly, or any Reaper / Logic / Audition session. The WAV files are unmodified PCM — no intermediate transcode, no format conversion.
Multi-track stems (WAV, per-guest)
Mixed-down stereo (WAV)
Mixed-down MP3 (320kbps)
Loudness-normalized master (-16 LUFS)
What Rebel Audio is not: a replacement for your Logic Pro, Adobe Audition, or Pro Tools post chain. We capture lossless local tracks per guest, run drift correction, and export clean BWF/WAV stems at 48kHz / 24-bit. Dialogue compression, de-essing, room-tone matching, and final mastering are yours to do in whatever DAW you use. We hand off the raw material — clean, synced, and loudness-targeted at -16 LUFS.
Ready to record your first episode?
No credit card needed. Free tier includes drift correction and local recording.