Voxbend

NEW IN 3.0 — SOUND PACKS Turn one take into a ZIP of chosen variations. Create four character recipes in one action, audition them, and batch-render the keepers as WAV or AAC. Each pack includes named audio, recipes, source/output hashes and measured output details. Your original remains unchanged. Up to 12 variations and 30 minutes combined output including tail allowance. Saved recipes are used; original audio is not included. Archives are unencrypted. Keep the take. Explore the voice. Voxbend turns recordings and imported audio into reusable sound-design sessions. Keep your original unchanged, shape a sound, and compare versions before sharing. SHAPE AND KEEP Start with 16 voice-effect presets or blend two recipes. Adjust pitch, independent speed, Vocal color EQ, reverb, distortion and ring modulation. Trim and fade without cutting the original. Save named variations, undo edits and return to a complete custom recipe later. PITCH LAB Inspect pitch evidence from the actual clip, select a region and choose among all 12 roots with chromatic, major, natural minor or pentatonic scales. Apply an intentional region or whole-clip transpose and hear the result. Scale actions shift a region uniformly from its median pitch; this is not live automatic tuning or individual-note correction. Analysis works best on a clear single pitched source and uses the first channel of stereo clips. COMPARE IN PLACE Seek and loop a selection. Switch between original and processed audio at the corresponding source position. Optional level matching attenuates the louder side using whole-clip RMS. MEASURED DELIVERY Render WAV at 16 or 24 bit, or AAC, at 44.1 or 48 kHz. Follow progress, cancel a render, and inspect the encoded output's duration, channels, sample peak and RMS before sharing. These are sample measurements, not LUFS or true-peak certification. Share readable recipe JSON separately; it includes settings and names, but no audio. ON YOUR DEVICE Record up to 60 seconds or import mono/stereo audio up to five minutes and 150 MB. Six synthetic reference clips let you explore without a microphone. No account or cloud audio processing. Microphone access is optional and used for recording. You choose any export destination. Vocal color is tonal EQ, not vocal-tract formant shifting or voice identification. Session deletion retains the original until you explicitly remove unused originals in Settings. Recovery files and copies outside the app are managed separately.