time_segmentation

time_segmentation(sig, fs, nperseg=2048, noverlap=None, is_ecma=False)

Function used for the segmentation of a time signal into smaller parts of audio (blocks).

Parameters:

• sig (numpy.array) – A 1-dimensional time signal array.

• fs (float) – The time signal sampling frequency.

• nperseg (int, optional) – Length of each segment. Defaults to 2048.

• noverlap (int, optional) – Number of points to overlap between segments. If None, noverlap = nperseg / 2. Defaults to None.

• is_ecma (bool, optional) – If is_ecma = True, the implementation is as described in Formula 16 (section 5.1.4) of ECMA-418-2. Defaults to False.

Returns:

• block_array (numpy.array) – A 2-dimensional array of size (nperseg, nseg) containing the segmented signal.

• time (numpy.array) – The time axis corresponding to the segmented signal, size (nseg,)