pyACA.NoveltyFlux Namespace Reference

Functions
	NoveltyFlux (X, f_s)
	computes the novelty measure per Spectral Flux

Function Documentation

NoveltyFlux()

NoveltyFlux	(		X,
			f_s )

computes the novelty measure per Spectral Flux

Parameters

X	spectrogram (dimension FFTLength X Observations)
f_s	sample rate of audio data

Returns

d_flux: novelty measure

Definition at line 12 of file NoveltyFlux.py.

def NoveltyFlux(X, f_s):
 
    isSpectrum = X.ndim == 1
    if isSpectrum:
        X = np.expand_dims(X, axis=1)
 
    # difference spectrum (set first diff to zero)
    X = np.c_[X[:, 0], X]
    afDeltaX = np.diff(X, 1, axis=1)
 
    # half-wave rectification
    afDeltaX[afDeltaX < 0] = 0
 
    # flux
    d_flux = np.sqrt((afDeltaX**2).sum(axis=0)) / X.shape[0]
 
    return np.squeeze(d_flux) if isSpectrum else d_flux