Skip to content

Presets

OpenVoxTuner has two distinct preset systems, both stored/serialized via the JUCE AudioProcessorValueTreeState (APVTS):

  1. Curve presets (factory + custom) — these drive the ovtdsp::PitchCurve in the Curve Editor.
  2. Plugin presets (factory + custom) — these store only APVTS parameter state, never the pitch curve.

Factory curve presets

The factory curve preset catalog is defined in ovtdsp::FactoryPresets.h / FactoryPresets.cpp and returned by getFactoryPresets(). Each entry is a FactoryPresetInfo { id, displayName, category, description }.

The list below shows the real id values as used by PitchCurve::loadPreset(id):

id Display name Category Description
default Default Basic Flat A3 reference tone.
robot_c3 Robot (C3) Robotic Monotone robotic voice at C3.
robot_c4 Robot (C4) Robotic Monotone robotic voice at C4.
spoken_male Spoken (Male) Vocal Character Natural spoken-male contour.
spoken_female Spoken (Female) Vocal Character Natural spoken-female contour.
bass Bass Voice Type Low male bass melody.
baritone Baritone Voice Type Baritone melody.
tenor Tenor Voice Type Tenor melody.
alto Alto Voice Type Alto melody.
mezzo Mezzo Voice Type Mezzo-soprano melody.
soprano Soprano Voice Type Soprano melody.

Legacy curve presets

Earlier documentation described five PitchCurve::loadPreset names — default, spoken, lyric, rap, robot. The current factory catalog has evolved into the gallery above (robot → robot_c3/robot_c4, spoken → spoken_male/spoken_female, plus per-voice-type melodic presets). The loadPreset() API still accepts preset keys for the gallery entries.

Factory plugin presets

These are built-in full-parameter presets (APVTS state only):

Preset Description
Default True factory defaults (all parameters at default).
Natural Light Subtle correction, transparent mode.
Modern Pop Fast correction, bright formant.
Ballad Slow Slow retarget, vibrato preserved.
Electronic Heavy correction, formant shift.
Live Vocal Balanced for live use.
Podcast Speech Optimized for spoken voice.

(Plugin presets store only APVTS parameter state, never the pitch curve.)

The Curve Editor exposes a browsable gallery (a modeless window of factory + custom presets with curve thumbnails / metadata), opened from the presetGalleryButton. applyFactoryPreset(name) resets the morph, loads the curve into the editor, commits it to the active A/B slot, and aligns the morph slider. A contextual menu (right-click in empty space, or the hamburger Options menu) also lists the presets.

Serialization & storage

  • APVTS: all parameters (key, scale, custom0…custom11, speed, amount, formant, harmony, reverb, noise gate, UI theme/language, etc.) are managed by the plugin's AudioProcessorValueTreeState. This synchronizes host ↔ GUI ↔ DSP and is serialized through getStateInformation() / setStateInformation().
  • Pitch curve: serialized as an XML sub-element <PITCH_CURVE> in getStateInformation(), via PitchCurve::toXml() / fromXml().
  • Custom presets: user-saved plugin presets are written to files (parameters.state only), while custom curve presets are saved/loaded separately and re-applied to the editor.

Custom scale state

When Scale::Custom (index 13) is active, the scale is defined by 12 AudioParameterBool (custom0…custom11, one per semitone 0=C … 11=B). The default is C major (C, D, E, F, G, A, B on; C#, D#, F#, G#, A# off). These 12 booleans are individually automatable by the host and are edited via the ScaleKeyboardComponent. The ScaleQuantizer consumes them through setCustomIntervals() (without key offset), and the Curve Editor's interactive snap uses the same custom notes.

See docs/default-parameters.md for the full table of parameters, including defaults, ranges, and the deprecated flex_tune / attack_aware / attack_release controls (kept for preset compatibility, DSP disabled since 2026-07-24).