Pairs and chords choose their own count. Groups append to the bank. Headphones separate binaural channels.
Stereo tests
Signal sequence
Set a tone's signal mode to Sequence. Each stage replaces that voice's waveform and frequency; routing, gain and modulation still apply.
Automation recipes
Load a recipe, then Start. Open the tone’s Automation section to edit it. Loading replaces the bank.
Bass drop: sine, 160 → 40 Hz over 3 s. Sweep: 40 → 1000 Hz over 10 s. Sub test: 20 → 100 Hz over 15 s. Sub drop: 80 → 25 Hz over 4 s. These use exponential frequency ramps with Loop off. Wobble: 60 Hz sine with amplitude 0.08 → 0.65, sine curve, 0.5 s cycle, Loop on (2 Hz).
For tempo sync, cycle seconds = 60 ÷ BPM ÷ cycles per beat. At 120 BPM, use 0.5 s for quarter notes or 0.25 s for eighth notes. Add a pan lane for stereo motion or automate modulation depth for changing intensity.
Save keeps the editable sound locally; Export JSON backs it up. Render WAV saves the generated audio, including automation. It renders from time zero; live MIDI knob movements change settings. Use Record live output to capture those movements; WAV renders the saved configuration.
MIDI controls
MIDI disconnected. Chrome/Edge recommended; access is requested only on Connect.
Connect, choose a target and control, Learn CC, move a knob, then Add mapping. MIDI channel and ranges are saved with presets. Press Start for audio. Automation takes priority over MIDI on the same control; remove that lane to use a knob.
Timer & transport fades
Timer repetitions restart the voice schedule; oscillator phase remains continuous. Rest silences the output.
Save audio
Live recording captures MIDI and manual changes as WebM/Opus, Ogg/Opus or M4A (browser dependent). Render WAV below saves uncompressed audio from time zero.
Offline render uses current configuration from time zero.
Pitch & frequency calculator
Diagnostics
Logs stay local. Enable with ?debug=1. No audio, account or analytics data is sent anywhere.