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.utils.preprocessing#

Description : Preprocessing utils implementation.
Copyright (c) 2019-2024 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>.

class spafe.utils.preprocessing.SlidingWindow(win_len: float = 0.025, win_hop: float = 0.01, win_type: typing_extensions.Literal[hanning, bartlet, kaiser, blackman, hamming] = 'hamming')[source]#

Bases: object

Sliding widow class.

Parameters

win_len (float) – window length in sec. (Default is 0.025).
win_hop (float) – step between successive windows in sec. (Default is 0.01).
win_type (float) – window type to apply for the windowing. (Default is “hamming”).

win_hop: float = 0.01#

win_len: float = 0.025#

win_type: typing_extensions.Literal[hanning, bartlet, kaiser, blackman, hamming] = 'hamming'#

spafe.utils.preprocessing.framing(sig: ndarray, fs: int = 16000, win_len: float = 0.025, win_hop: float = 0.01) → Tuple[ndarray, int][source]#

transform a signal into a series of overlapping frames (= Frame blocking) as described in [Malek-framing-blog].

Parameters

sig (numpy.ndarray) – a mono audio signal (Nx1) from which to compute features.
fs (int) – the sampling frequency of the signal we are working with. (Default is 16000).
win_len (float) – window length in sec. (Default is 0.025).
win_hop (float) – step between successive windows in sec. (Default is 0.01).

Returns

(numpy.ndarray) : array of frames.
(int) : frame length.

Return type

(tuple)

Note

Uses the stride trick to accelerate the processing.

References

Malek-framing-blog: : Malek A., Signal framing, 25.01.2022, https://superkogito.github.io/blog/2020/01/25/signal_framing.html

spafe.utils.preprocessing.pre_emphasis(sig: ndarray, pre_emph_coeff: float = 0.97) → ndarray[source]#

perform preemphasis on the input signal.

Parameters

sig (numpy.ndarray) – input signal.
coeff (float) – preemphasis coefficient. 0 is no filter. (Default is 0.97).

Returns

pre-empahsised signal.

Return type

(numpy.ndarray)

Note

\[y[t] = x[t] - \alpha \times x[t-1]\]

spafe.utils.preprocessing.stride_trick(a: ndarray, stride_length: int, stride_step: int) → ndarray[source]#

apply framing using the stride trick from numpy.

Parameters

a (numpy.ndarray) – signal array.
stride_length (int) – length of the stride.
stride_step (int) – stride step.

Returns

blocked/framed array.

Return type

(numpy.ndarray)

Note

You can refer to numpy documentation of this stride trick here: https://numpy.org/doc/stable/reference/generated/numpy.lib.stride_tricks.sliding_window_view.html

spafe.utils.preprocessing.windowing(frames: ndarray, frame_len: int, win_type: typing_extensions.Literal[hanning, bartlet, kaiser, blackman, hamming] = 'hamming') → ndarray[source]#

generate and apply a window function to avoid spectral leakage [Malek-windowing-blog].

Parameters

frames (numpy.ndarray) – array including the overlapping frames.
frame_len (int) – frame length.
win_type (str) – type of window to use. (Default is “hamming”).

Returns

windowed frames.

Return type

(numpy.ndarray)

References

Malek-windowing-blog: : Malek, A. Specctral leakage, 2022.03.13, https://superkogito.github.io/blog/2020/03/13/spectral_leakage_and_windowing.html

spafe.utils.preprocessing.zero_handling(x: ndarray) → ndarray[source]#

handle the issue with zero values if they are exposed to become an argument for any log function.

Parameters: x (numpy.ndarray) – input vector.
Returns: vector with zeros substituted with epsilon values.
Return type: (numpy.ndarray)

spafe.utils.filters

spafe.utils.spectral