Skip to content

Audio Analysis

OpenScofo uses audio descriptors for onset detection, pitch tracking, and AI models.

ONNXMODEL "flute.onnx"
ONNXDESCRIPTORS mfcc zcr centroid spread

PTECH pizz C4 1
    sendto sample [start pizz_echo]

flowchart LR
    audio[Live audio] --> descriptors[Descriptors]
    descriptors --> model[Recognition model]
    model --> cue[Score event]

Reference Table

Area Use Reference
Amplitude loudness, RMS, dB, silence Amplitude Descriptors
Onset attack detection Onset Descriptor
Time / pitch ZCR, pitch, confidence Time-Domain and Pitch Descriptors
Spectral timbre, brightness, MFCC, chroma Spectral Descriptors

Online descriptor test

Test OpenScofo descriptors.

Remarks

Compatibility icons:

  • image/svg+xml compatible with librosa
  • compatible with essentia

See also: AI Models, Advanced Configuration.