We use cookies on this website to distinguish you from other users. We use this data to improve our content experience and for targeted advertising. By continuing to use this website you consent to our use of cookies. For more information, please see our Cookie Policy.

spafe.fbanks.bark_fbanks#

Description : Bark filter banks implementation.
Copyright (c) 2019-2023 Ayoub Malek. This source code is licensed under the terms of the BSD 3-Clause License. For a copy, see <https://github.com/SuperKogito/spafe/blob/master/LICENSE>.

spafe.fbanks.bark_fbanks.Fm(fb: float, fc: float) → float[source]#

Compute a Bark filter around a certain center frequency in bark [Hermansky].

Parameters

fb (float) – frequency in Bark.
fc (float) – center frequency in Bark.

Returns

associated Bark filter value/amplitude.

Return type

(float)

spafe.fbanks.bark_fbanks.bark_filter_banks(nfilts: int = 24, nfft: int = 512, fs: int = 16000, low_freq: float = 0, high_freq: Optional[float] = None, scale: Literal['ascendant', 'descendant', 'constant'] = 'constant', conversion_approach: Literal['Wang', 'Tjomov', 'Schroeder', 'Terhardt', 'Zwicker', 'Traunmueller'] = 'Wang')[source]#

Compute Bark filter banks. The filters are stored in the rows, the columns correspond to fft bins.

Parameters

nfilts (int) – the number of filters in the filter bank. (Default is 20).
nfft (int) – the FFT size. (Default is 512).
fs (int) – sample rate/ sampling frequency of the signal. (Default 16000 Hz).
low_freq (float) – lowest band edge of mel filters. (Default 0 Hz).
high_freq (float) – highest band edge of mel filters. (Default is fs/2).
scale (str) – monotonicity behavior of the filter banks. (Default is “constant”).
conversion_approach (str) – bark scale conversion approach. (Default is “Wang”).

Returns

(numpy.ndarray) : array of size nfilts * (nfft/2 + 1) containing filter bank. Each row holds 1 filter.
(numpy.ndarray) : array of center frequencies

Return type

(tuple)

Raises

ParameterError –

if low_freq < 0 OR high_freq > (fs / 2)

Tip

scale : can take the following options [“constant”, “ascendant”, “descendant”].
conversion_approach : can take the following options [“Tjomov”,”Schroeder”, “Terhardt”, “Zwicker”, “Traunmueller”, “Wang”]. Note that the use of different options than the ddefault can lead to unexpected behavior/issues.

References

Hermansky: Hermansky, H. “Perceptual linear predictive (PLP) analysis of speech.” The Journal of the Acoustical Society of America 87 4 (1990): 1738-52 doi: 10.1121/1.399423. PMID: 2341679.

Examples

import numpy as np
from spafe.utils.converters import bark2hz
from spafe.utils.vis import show_fbanks
from spafe.fbanks.bark_fbanks import bark_filter_banks

# init var
fs = 8000
nfilt = 7
nfft = 1024
low_freq = 0
high_freq = fs / 2

# compute freqs for xaxis
bhz_freqs = np.linspace(low_freq, high_freq, nfft //2+1)

for scale, label in [("constant", ""), ("ascendant", "Ascendant "), ("descendant", "Descendant ")]:
    # bark fbanks
    bark_fbanks_mat, bark_freqs = bark_filter_banks(nfilts=nfilt,
                                                    nfft=nfft,
                                                    fs=fs,
                                                    low_freq=low_freq,
                                                    high_freq=high_freq,
                                                    scale=scale)

    # visualize filter bank
    show_fbanks(
        bark_fbanks_mat,
        [bark2hz(freq) for freq in bark_freqs],
        bhz_freqs,
        label + "Bark Filter Bank",
        ylabel="Weight",
        x1label="Frequency / Hz",
        x2label="Frequency / bark",
        figsize=(14, 5),
        fb_type="bark",
    )