pyACA.FeatureSpectralCentroid Namespace Reference

Functions
	FeatureSpectralCentroid (X, f_s)
	computes the spectral centroid from the (squared) magnitude spectrum

Function Documentation

FeatureSpectralCentroid()

FeatureSpectralCentroid	(		X,
			f_s )

computes the spectral centroid from the (squared) magnitude spectrum

Parameters

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

Returns

vsc: spectral centroid (in Hz)

Definition at line 12 of file FeatureSpectralCentroid.py.

def FeatureSpectralCentroid(X, f_s):
 
    isSpectrum = X.ndim == 1
 
    # X = X**2 removed for consistency with book
 
    norm = X.sum(axis=0, keepdims=True)
    norm[norm == 0] = 1
 
    vsc = np.dot(np.arange(0, X.shape[0]), X) / norm
 
    # convert from index to Hz
    vsc = vsc / (X.shape[0] - 1) * f_s / 2
 
    # if input is a spectrum, output scaler else if spectrogram, output 1d array
    vsc = np.squeeze(vsc) if isSpectrum else np.squeeze(vsc, axis=0)
 
    return vsc