calpy.utilities package

Submodules

calpy.utilities.utilities module

calpy.utilities.utilities.artificial_signal(frequencys, sampling_frequency=16000, duration=0.025)

Concatonates a sequence of sinusoids of frequency f in frequencies.

Args:

frequencys (list-like(int)): sequence of frequencies of sinusoidual signals in Hz. sampling_frequency (int, optional): sampling frequency in Hz. Defaults to 16000. duration (float, optional): duration of the output sinusoid in seconds. Defaults to 0.025.

Returns:

numpy.array: artificially generated sinusoidal signal.

calpy.utilities.utilities.compress_pause_to_time(signal, sampling_rate, time_step=0.01, frame_window=0.025)

compress pause index to time Args:

signal (numpy.array(bool)): A list of pause sequence. True indicating pause. sampling_rate (int): sampling frequency in Hz. time_step (float, optional): The time interval (in seconds) between two pauses. Default to 0.01. frame_window (float, optional): The length of speech (in seconds) used to estimate pause. Default to 0.025.

Returns:

numpy.array(bool): compressed pause.

calpy.utilities.utilities.is_lower_triangular(AA)

Check if a matrix is lower triangular. Args:

AA (numpy.array): a 2D matrix.

Returns: bool:

calpy.utilities.utilities.is_upper_triangular(AA)

Check if a matrix is upper triangular. Args:

AA (numpy.array): a 2D matrix.

Returns: bool:

calpy.utilities.utilities.merge_pitch_profile(pitches, speaker_id)

merges n-pitch profiles into one sound based on speaker_id.

Args:

pitches (list-like(float)): a sequence of pitches. speaker_id (list-like(int)): a list of speakers’ id.

Returns:

numpy.array: merged pitch profile.

calpy.utilities.utilities.pad_signal(signal, sampling_rate, time_step=0.01, frame_window=0.025)

segement a signal, 1D audio signal, into frames, such that: output: N by M matrix, in which:

each row is a segment of frame_window’s audio signal

calpy.utilities.utilities.random_run(distributions, length, min_run=100, max_more=100)

calpy.utilities.utilities.random_symbols(distribution, length)

calpy.utilities.utilities.read_wavfile(filename, channel=0)

Read in a audio file (in .wav format) and enforce the output as mono-channel. Args:

filename (str): path to the audio file. channel(int, optional): indicate which channel to read in. Defaults to 0.

Returns:

int: sampling frequency. numpy.array: audio data.

calpy.utilities.utilities.sinusoid(frequency, sampling_frequency=16000, duration=0.025)

Generate a sinusoid signal. Args:

frequency (int): the frequency of the sinusoidal signal. sampling_frequency (int, optional): sampling frequency in Hz. Defaults to 16000. duration (float, optional): duration of the output sinusoid in seconds. Defaults to 0.025.

Returns: numpy.array: a sinusoid.

calpy.utilities.utilities.write_wavfile(filename, fs, data)