clarity.evaluator.msbg.smearing module

Temporal smearing component of MSBG model.

class clarity.evaluator.msbg.smearing.Smearer(rl: float, ru: float, sample_rate: float)[source]

Bases: object

Class to hold the re-usable smearing filter.

smear(input_signal: ndarray) → ndarray[source]

Smear a given input signal.

clarity.evaluator.msbg.smearing.audfilt(rl: float, ru: float, sample_rate: float, asize: int = 256) → ndarray[source]

Calculate an auditory filter array.

Parameters:

  • rl (float) – broadening factor on the lower side

  • ru (float) – broadening factor on the upper side

  • sample_rate (float) – signal sampling frequency

  • asize (int, optional) – number of taps in filter (default: {256})

Returns:

A bank of auditory filters stored as 2-d numpy array

Return type:

ndarray

clarity.evaluator.msbg.smearing.make_smear_mat3(rl: float, ru: float, sample_rate: float) → ndarray[source]

Make the smearing filter matrix.

Parameters:

  • rl (float) – filter broadening factor on the lower side

  • ru (float) – filter broadening factor on the upper side

  • sample_rate (float) – sampling frequency

Returns:

The FFT_SIZE/2 X FFT_SIZE/2 smearing filter matrix

Return type:

ndarray

clarity.evaluator.msbg.smearing.smear3(f_smear: ndarray, inbuffer: ndarray) → ndarray[source]

Direct translation of smear3.m from MSBG hearing loss model.

Parameters:

  • f_smear (ndarray) – The FFT_SIZE/2 X FFT_SIZE/2 smearing filter matrix

  • inbuffer (ndarray) – signal with prepended tone and noise

Returns:

outbuffer

Return type:

ndarray