librosa.filters.window_bandwidth

librosa.filters.window_bandwidth(window, n=1000)[source]

Get the equivalent noise bandwidth (ENBW) of a window function.

The ENBW of a window is defined by [1] (equation 11) as the normalized ratio of the sum of squares to the square of sums:

enbw = n * sum(window**2) / sum(window)**2

Parameters:

windowcallable or string: A window function, or the name of a window function, e.g.: scipy.signal.hann or ‘boxcar’
nint > 0: The number of coefficients to use in estimating the window bandwidth

Returns:

bandwidthfloat: The equivalent noise bandwidth (in FFT bins) of the given window function