Pitch down a fifth, vocal tract enlarged. Sounds like a much larger speaker.
One speaker. Add a voice to make it several.
Records your voice and transforms it with a phase vocoder, which is what allows pitch and vocal tract size to move independently. Lowering pitch alone gives the familiar deepening; moving the formants without touching pitch makes the speaker seem the wrong size, which is the more unsettling result.
It also offers time stretching without pitch change, whisperisation that removes voicing entirely, layering into several speakers from a single take, plus breath, space and ring modulation. Audio is processed locally and exported as a WAV.
Start by lowering the vocal tract slightly without changing pitch, add a little time stretch, then breath and space. Whispering removes the voiced tone entirely and works well for distant, breathy voices.
A technique that splits sound into many narrow frequency bands over short time windows, so timing, pitch and the spectral envelope can be changed separately before the sound is rebuilt.
No. Recording and processing happen in your browser and the WAV is saved directly to your device.
No. It transforms a recording rather than a live microphone. That is what allows the higher-quality processing and the A/B comparison.
This is an instrument and a sound tool, provided for entertainment and experiment. It describes what it measures and does not identify the cause of anything it shows.