Gammatone Filter Bank¶

Examples¶

TODO:

• Tests,
• nice introduction with example,

References¶

Functions¶

class gammatone.GammatoneFilterbank(samplerate=44100, order=4, startband=-12, endband=12, normfreq=1000.0, density=1.0, bandwidth_factor=1.0, desired_delay_sec=0.02)[source]¶

Returns a GammatoneFilterbank instance for filtering signals.

Parameters:	samplerate : scalar Default: 44100. order : scalar Default: 4. startband : scalar Default: -12, endband : scalar Default: 12, normfreq : scalar Default: 1000.0, density : scalar Default: 1.0, bandwidth_factor : scalar Default: 1.0, desired_delay_sec : scalar Default: 0.02

Methods

analyze(signal, states=None)[source]¶

Returns a generator yielding filtered signal bands and states.

delay(bands)[source]¶

Returns delayed bands for an optimized impulseresponse at synthesis.

estimate_max_indices_and_slopes(delay_samples=None)[source]¶

Returns Estimate of maximum index and slopes at max. Used for estimation of phase-factors for delaying bands.

freqz(nfft=4096, plotfun=None)[source]¶

Returns frequency and impulse responses. Accepts plotfun function for cusomized plotting.

init_delay(desired_delay_sec)[source]¶

Initializes delay estimation and variables for delay bands to achieve a optimized impulse response.

init_gains()[source]¶

Initializes gains for weighting bands leading to a equalized freqresponse when summing (synthesizing) bands.

reanalyze(bands, states=None)[source]¶

Refilters already filtered signal bands if bandwidening effect of a used algorithm needs to be reduced.

synthesize(bands)[source]¶

Returns summed dleayed and weighted bands.

gammatone.design_filtbank_coeffs(samplerate, order, centerfrequencies, bandwidths=None, bandwidth_factor=None, attenuation_half_bandwidth_db=-3)[source]¶

Designs complex coefficients for a gammatone filter bank.

Parameters:	samplerate : int order : int centerfrequencies : arraylike bandwidths : arraylike (optional) bandwidth_factor : scalar (1.0 is auditory erb) attenuation_half_bandwidth_db : scalar
Returns:	generator of b, a coefficients

gammatone.design_filter(sample_rate=44100, order=4, centerfrequency=1000.0, band_width=None, band_width_factor=1.0, attenuation_half_bandwidth_db=-3)[source]¶

Returns filter coefficient of a gammatone filter [Hohmann2002].

Parameters:	sample_rate : int/scalar order : int centerfrequency : scalar band_width : scalar band_width_factor : scalar attenuation_half_bandwidth_db : scalar
Returns:	b, a : ndarray, ndarray

gammatone.erb_aud(centerfrequency)[source]¶

Retrurns equivalent rectangular band width of an auditory filter. Implements Equation 13 in [Hohmann2002].

Parameters:	centerfrequency : scalar /Hz The center frequency in Hertz of the desired auditory filter.
Returns:	erb : scalar Equivalent rectangular bandwidth of an auditory filter at centerfrequency.

gammatone.erb_count(centerfrequency)[source]¶

Returns the equivalent rectangular band count up to centerfrequency.

Parameters:	centerfrequency : scalar /Hz The center frequency in Hertz of the desired auditory filter.
Returns:	count : scalar Number of equivalent bandwidths below centerfrequency.

gammatone.erbscale_to_hertz(erb)[source]¶

Returns frequency in Hertz from ERB value. Implements Equation 17 in [Hohmann2002].

Parameters:	erb : scalar The corresponding value on the ERB-Scale.
Returns:	frequency : scalar The Frequency in Hertz.

gammatone.fosfilter(b, a, order, signal, states=None)[source]¶

Return signal filtered with b and a (first order section) by filtering the signal order times.

This Function was created for filtering signals by first order section cascaded complex gammatone filters.

Parameters:	b, a : ndarray, ndarray Filter coefficients of a first order section filter. Can be complex valued. order : int Order of the filter to be applied. This will be the count of refiltering the signal order times with the given coefficients. signal : ndarray Input signal to be filtered. states : ndarray, default None Array with the filter states of length order. Initial you can set it to None.
Returns:	signal : ndarray Output signal, that is filtered and complex valued (analytical signal). states : ndarray Array with the filter states of length order. You need to loop it back into this function when block processing.

gammatone.frequencies_gammatone_bank(start_band, end_band, norm_freq, density)[source]¶

Returns centerfrequencies and auditory Bandwidths for a range of gamatone filters.

Parameters:	start_band : int Erb counts below norm_freq. end_band : int Erb counts over norm_freq. norm_freq : scalar The reference frequency where all filters are around density : scalar ERB density 1would be erb_aud.
Returns:	centerfrequency_array : ndarray

gammatone.freqz_fos(b, a, order, nfft, plotfun=None)[source]¶

Returns Frequency Response and impulse response of first order section coeffs.

Parameters:	b : arraylike a : arraylike order : int nfft : int plotfun : function(frequencies, response)
Returns:	freqresponse : arraylike frequencies : arraylike response : arraylike

gammatone.hertz_to_erbscale(frequency)[source]¶

Returns ERB-frequency from frequency in Hz. Implements Equation 16 in [Hohmann2002].

Parameters:	frequency : scalar The Frequency in Hertz.
Returns:	erb : scalar The corresponding value on the ERB-Scale.