:py:mod:`tonic.audio_transforms` ================================ .. py:module:: tonic.audio_transforms Module Contents --------------- Classes ~~~~~~~ .. autoapisummary:: tonic.audio_transforms.FixLength tonic.audio_transforms.Bin tonic.audio_transforms.SwapAxes tonic.audio_transforms.AmplitudeScale tonic.audio_transforms.RobustAmplitudeScale tonic.audio_transforms.SOSFilter tonic.audio_transforms.ButterFilter tonic.audio_transforms.ButterFilterBank tonic.audio_transforms.LinearButterFilterBank tonic.audio_transforms.MelButterFilterBank tonic.audio_transforms.AddNoise Functions ~~~~~~~~~ .. autoapisummary:: tonic.audio_transforms.normalize .. py:class:: FixLength Fix the length of a sample along a specified axis to a given length. :param length: Desired length of the sample :type length: int :param axis: Dimension along which the length needs to be fixed.. Defaults to 1. :type axis: int, optional :param data: data sample :type data: np.ndarray :returns: fixed length data sample :rtype: np.ndarray .. py:attribute:: length :type: int .. py:attribute:: axis :type: int :value: 1 .. py:method:: __call__(data: numpy.ndarray) -> numpy.ndarray .. py:class:: Bin Bin the given data along a specified axis at the specified new frequency. :param orig_freq: Sampling frequency of the given data stream :type orig_freq: float :param new_freq: Desired frequency after binning :type new_freq: float :param axis: Axis along which the data needs to be binned :type axis: int :param data: data sample :type data: np.ndarray :returns: binned data sample :rtype: np.ndarray .. py:attribute:: orig_freq :type: float .. py:attribute:: new_freq :type: float .. py:attribute:: axis :type: int .. py:method:: __call__(data: numpy.ndarray) -> numpy.ndarray .. py:class:: SwapAxes Interchange two axes of an array. :param ax1: First axis :type ax1: int :param ax2: Second axis :type ax2: int :returns: array with interchanged axes :rtype: np.ndarray .. py:attribute:: ax1 :type: int :value: 0 .. py:attribute:: ax2 :type: int :value: 1 .. py:method:: __call__(data: numpy.ndarray) -> numpy.ndarray .. py:class:: AmplitudeScale Normalize the maximum amplitude of the input signal to a desired value, for instance the value can corespond to the mVolt equivalent of be the maximum loudness of the audio. :param data: input single- or multi-channel signal. :type data: np.ndarray :returns: scaled version of the signal. :rtype: np.ndarray .. py:attribute:: max_amplitude :type: float :value: 0.158 .. py:method:: __call__(data: numpy.ndarray) -> numpy.ndarray .. py:class:: RobustAmplitudeScale Normalize the maximum amplitude of the input signal while eliminating the outliers, to ensure that some very large outlier samples do not attenuate the majority of the samples. Paremeters: max_robust_amplitude (float): maximum amplitude of non-outlier samples of the signal after robust scaling. outlier_percent (float, optional): percentage of the outlier in the data. Defaults to 0.01. :param data: input single- or multi-channel signal. :type data: np.ndarray :returns: scaled version of the signal. :rtype: np.ndarray .. py:attribute:: max_robust_amplitude :type: float :value: 0.158 .. py:attribute:: outlier_percent :type: float :value: 0.01 .. py:method:: __call__(data: numpy.ndarray) -> numpy.ndarray .. py:class:: SOSFilter SOS filter. :param coeffs: coefficients of the second order filter :param axis: Axis along with the filter needs to be applied :param See https: :type See https: //docs.scipy.org/doc/scipy/reference/generated/scipy.signal.sosfilt.html for more details .. py:attribute:: coeffs :type: numpy.ndarray .. py:attribute:: axis :type: int .. py:method:: __call__(signal) .. py:class:: ButterFilter Butter filter. :param order: Order of filter to be used :param freq: Frequency for the filter (float or (float, float)) :param analog: True if analog filter :param btype: Filter type, {‘lowpass’, ‘highpass’, ‘bandpass’, ‘bandstop’} :param rectify: If true, the output is the absolute value of the filtered output :param axis: Axis along which the filter needs to be applied :param See https: :type See https: //docs.scipy.org/doc/scipy/reference/generated/scipy.signal.butter.html#scipy.signal.butter for more details on parameters. .. py:attribute:: order :type: int .. py:attribute:: freq :type: float | tuple[float, float] .. py:attribute:: analog :type: bool .. py:attribute:: btype :type: str .. py:attribute:: rectify :type: bool .. py:attribute:: axis :type: int .. py:method:: __post_init__() .. py:method:: __call__(data: numpy.ndarray) -> numpy.ndarray .. py:class:: ButterFilterBank Butter filter bank. :param order: Order of filter to be used :param freq: Frequency for the filter (float or (float, float)) :param rectify: If true, the output is the absolute value of the filtered output :param axis: Axis along which the filter needs to be applied :param analog: If true, the filter will be analog. False by default :param See https: :type See https: //docs.scipy.org/doc/scipy/reference/generated/scipy.signal.butter.html#scipy.signal.butter for more details on parameters. .. py:attribute:: order :type: int .. py:attribute:: freq :type: list[tuple[float, float]] .. py:attribute:: rectify :type: bool .. py:attribute:: axis :type: int .. py:attribute:: analog :type: bool :value: False .. py:method:: __post_init__() .. py:method:: __call__(data) .. py:class:: LinearButterFilterBank Butter filter bank. :param order: Order of filter to be used :param low_freq: Lower/cutoff frequency the filter (float or (float, float)) :param sampling_freq: Sampling frequency of the signal, also serves as higher frequency of the filter bank. :param analog: True if analog filter :param rectify: If true, the output is the absolute value of the filtered output :param axis: Axis along which the filter needs to be applied :param See https: :type See https: //docs.scipy.org/doc/scipy/reference/generated/scipy.signal.butter.html#scipy.signal.butter for more details on parameters. .. py:attribute:: order :type: int :value: 2 .. py:attribute:: low_freq :type: float :value: 100 .. py:attribute:: sampling_freq :type: float :value: 16000 .. py:attribute:: analog :type: bool :value: False .. py:attribute:: num_filters :type: int :value: 64 .. py:attribute:: rectify :type: bool :value: True .. py:attribute:: axis :type: int .. py:method:: compute_freq_bands() .. py:method:: __post_init__() .. py:method:: __call__(data) .. py:class:: MelButterFilterBank Bases: :py:obj:`LinearButterFilterBank` Butter filter bank with frequencies along the mel scale. :param order: Order of filter to be used :param low_freq: Lower/cutoff frequency the filter (float or (float, float)) :param sampling_freq: Sampling frequency of the signal, also serves as higher frequency of the filter bank. :param analog: True if analog filter :param rectify: If true, the output is the absolute value of the filtered output :param axis: Axis along which the filter needs to be applied :param See https: :type See https: //docs.scipy.org/doc/scipy/reference/generated/scipy.signal.butter.html#scipy.signal.butter for more details on parameters. .. py:method:: hz2mel(freq) :staticmethod: .. py:method:: mel2hz(freq) :staticmethod: .. py:method:: compute_freq_bands() .. py:class:: AddNoise Add nose to data. Params: dataset: A dataset object that returns a tuple when iterated over the first element of which is the audio signal to be used for noise. snr: Desired signal to noise ratio in dB normed: If set to false, the signal max value will not be normalized. True by default. .. py:attribute:: dataset :type: collections.abc.Iterator .. py:attribute:: snr :type: float .. py:attribute:: normed :type: bool :value: True .. py:method:: get_noise_sample(sample_len: int) -> numpy.ndarray Get a random noise sample from the dataset. .. py:method:: __call__(signal) .. py:function:: normalize(signal) Normalize the signal.