aug module¶

This module defines class for augmentation of MultiChannelSignal objects and WAV audio flies.

class aug.SignalAugmentation(signal: MultiChannelSignal = None, seed=-1)¶

Class provides augmentation of multichannel sound data (MultiChannelSignal object) and audio files.

achn(list_of_chains: list[str]) → MultiChannelSignal¶

Add chain to list of chains.

Parameters:

list_of_chains (list[str]) – A list of chains to add.

Returns:

The updated SignalAugmentation instance: with added chains. Result, allowing for method chaining.

Return type:

self (SignalAugmentation)

add_chain(list_of_chains: list[str]) → MultiChannelSignal¶

Add chain to list of chains.

Parameters:

list_of_chains (list[str]) – A list of chains to add.

Returns:

The updated SignalAugmentation instance: with added chains. Result, allowing for method chaining.

Return type:

self (SignalAugmentation)

amp(amplitude_list: list[float], amplitude_deviation_list: list[float] = None) → SignalAugmentation¶

Apply amplitude control to a multichannel sound. If amplitude_deviation_list is defined, you can get different versions of the same MultiChannelSignal data.

Parameters:

amplitude_list (list[float]) – The list of amplitude coefficients to apply to each channel.
amplitude_deviation_list (list[float]) – If exists, sets amplitude values random with uniform distribution in range [amplitude - deviation, amplitude + deviation)].

Returns:

Object with amplitude-controlled: multichannel sound.

Return type:

self (SignalAugmentation)

amplitude_ctrl(amplitude_list: list[float], amplitude_deviation_list: list[float] = None) → SignalAugmentation¶

Apply amplitude control to a multichannel sound. If amplitude_deviation_list is defined, you can get different versions of the same MultiChannelSignal data.

Parameters:

amplitude_list (list[float]) – The list of amplitude coefficients to apply to each channel.
amplitude_deviation_list (list[float]) – If exists, sets amplitude values random with uniform distribution in range [amplitude - deviation, amplitude + deviation)].

Returns:

Object with amplitude-controlled: multichannel sound.

Return type:

self (SignalAugmentation)

copy() → SignalAugmentation¶: Deep copy of the SignalAugmentation object.

cpy() → SignalAugmentation¶: Deep copy of the SignalAugmentation object.

delay_ctrl(delay_us_list: list[int], delay_deviation_list: list[int] = None) → SignalAugmentation¶

Add delays of channels of multichannel sound. Output data become longer. Values of delay will be converted to count of samples.

Parameters:

delay_us_list (list[int]) – The list of delay values in microseconds
integer. (to apply to each channel. Each value should be a positive)
data. (sound)
delay_deviation_list (list[int]) – If exists, the list of delay
distributed. (deviations makes delays uniformly)

Returns:

Object with delayed multichannel sound.

Return type:

self (SignalAugmentation)

dly(delay_us_list: list[int], delay_deviation_list: list[int] = None) → SignalAugmentation¶

Add delays of channels of multichannel sound. Output data become longer. Values of delay will be converted to count of samples.

Parameters:

delay_us_list (list[int]) – The list of delay values in microseconds
integer. (to apply to each channel. Each value should be a positive)
data. (sound)
delay_deviation_list (list[int]) – If exists, the list of delay
distributed. (deviations makes delays uniformly)

Returns:

Object with delayed multichannel sound.

Return type:

self (SignalAugmentation)

echo(delay_us_list: list[int], amplitude_list: list[float], delay_deviation_list: list[int] = None, amplitude_deviation_list: list[float] = None) → SignalAugmentation¶

Add echo to multichannel sound. The output data become longer. To each channel will be added it’s copy with corresponding delay delay and amplitude. It looks like acoustic wave was reflected from the hard wall.

Parameters:

delay_us_list (list[int]) – The list of delay values in microseconds to apply to each channel. Each value should be a positive integer.
amplitude_list (list[float]) – The list of amplitude coefficients to apply to each channel.
delay_deviation_list (list[int]) – If exists gives random deviation of reflection delay.
amplitude_deviation_list (list[float]) – If exists gives random deviation of reflection amplitude.

Returns:

The object with echoed multichannel: sound.

Return type:

self (SignalAugmentation)

echo_ctrl(delay_us_list: list[int], amplitude_list: list[float], delay_deviation_list: list[int] = None, amplitude_deviation_list: list[float] = None) → SignalAugmentation¶

Add echo to multichannel sound. The output data become longer. To each channel will be added it’s copy with corresponding delay delay and amplitude. It looks like acoustic wave was reflected from the hard wall.

Parameters:

delay_us_list (list[int]) – The list of delay values in microseconds to apply to each channel. Each value should be a positive integer.
amplitude_list (list[float]) – The list of amplitude coefficients to apply to each channel.
delay_deviation_list (list[int]) – If exists gives random deviation of reflection delay.
amplitude_deviation_list (list[float]) – If exists gives random deviation of reflection amplitude.

Returns:

The object with echoed multichannel: sound.

Return type:

self (SignalAugmentation)

eval() → list[MultiChannelSignal]¶

Evaluate list of chains.

Parameters:: none
Returns:: The updated SignalAugmentation instance with added chains. Result, allowing for method chaining.
Return type:: self (SignalAugmentation)

gen(frequency_list: list[int], duration: float = 5, sampling_rate: int = -1, mode='sine') → SignalAugmentation¶

Generates a multichannel sound based on the given frequency list, duration, sample rate, and mode.

Parameters:

frequency_list (list[int]) – A list of frequencies to generate sound
duration (for.) – The duration of the sound in seconds.
ms.DEF_SIGNAL_LEN. (Defaults to)
sampling_rate (int) – The sample rate of the sound. Defaults to -1.
mode (str) – The mode of sound generation. Can be ‘sine’ or
the ('speech'. Defaults to 'sine'. See details in)
MultiChannelSignal.

Returns:

Object with generated multichannel sound.

Return type:

self (SignalAugmentation)

generate(frequency_list: list[int], duration: float = 5, sampling_rate: int = -1, mode='sine') → SignalAugmentation¶

Generates a multichannel sound based on the given frequency list, duration, sample rate, and mode.

Parameters:

frequency_list (list[int]) – A list of frequencies to generate sound
duration (for.) – The duration of the sound in seconds.
ms.DEF_SIGNAL_LEN. (Defaults to)
sampling_rate (int) – The sample rate of the sound. Defaults to -1.
mode (str) – The mode of sound generation. Can be ‘sine’ or
the ('speech'. Defaults to 'sine'. See details in)
MultiChannelSignal.

Returns:

Object with generated multichannel sound.

Return type:

self (SignalAugmentation)

get() → MultiChannelSignal¶

Returns the multichannel sound data stored in the MultiChannelSignal instance.

Returns:: The multichannel sound data.
Return type:: signal (MultiChannelSignal)

info() → dict¶

Returns a dictionary containing information about the SignalAugmentation object.

The dictionary includes the information returned by the info() method of the signal attribute of the object, as well as a key-value pair where the key is ‘chains’ and the value is the chains attribute of the object.

Returns:: A dictionary containing information about the object.
Return type:: dict

merge() → SignalAugmentation¶

Mixes channels of a multichannel sound into a single channel.

Parameters:

none

Returns:

The object with merged sound data,: containing a single channel.

Return type:

self (SignalAugmentation)

mrg() → SignalAugmentation¶

Mixes channels of a multichannel sound into a single channel.

Parameters:

none

Returns:

The object with merged sound data,: containing a single channel.

Return type:

self (SignalAugmentation)

noise_ctrl(noise_level_list: list[float]) → SignalAugmentation¶

Apply noise to a multichannel sound.

Parameters:

noise_level_list (list[float]) – The list of noise levels to apply to each channel.

Returns:

The object noise-controlled multichannel: sound.

Return type:

self (SignalAugmentation)

ns(noise_level_list: list[float]) → SignalAugmentation¶

Apply noise to a multichannel sound.

Parameters:

noise_level_list (list[float]) – The list of noise levels to apply to each channel.

Returns:

The object noise-controlled multichannel: sound.

Return type:

self (SignalAugmentation)

pause_detect(relative_level: list[float]) → ndarray[int]¶

Detects pauses in a multichannel sound.

Args: relative_level (list[float]): The list of relative levels for each channel, signal below this level will be marked as pause.

Returns:

The mask indicating the pauses in the multichannel: sound. The mask has the same shape as the input sound. It contains zeros and ones 0 - pause, 1 - not a pause.

Return type:

np.ndarray

pause_set(pause_map: list, pause_sz: list[int]) → SignalAugmentation¶

Set pauses lengths in multichannel sound to selected values.

Parameters:

pause_map (list) – A list of dictionaries containing pairs of (index, length) of pauses for each channel.
pause_sz (list[int]) – A list of pause sizes (in samples) for each channel.

Returns:

The object with multichannel sound: with pauses shrunk.

Return type:

self (SignalAugmentation)

pause_shrink(mask: ndarray[int], min_pause: list[int]) → SignalAugmentation¶

Shrink pauses in multichannel sound.

Parameters:

mask (np.ndarray) – The mask indicating the pauses in the multichannel sound.
min_pause (list[int]) – The list of minimum pause lengths for each channel in samples.

Returns:

The object with multichannel sound with: pauses shrunk.

Return type:

self (SignalAugmentation)

pdt(relative_level: list[float]) → ndarray[int]¶

Detects pauses in a multichannel sound.

Args: relative_level (list[float]): The list of relative levels for each channel, signal below this level will be marked as pause.

Returns:

The mask indicating the pauses in the multichannel: sound. The mask has the same shape as the input sound. It contains zeros and ones 0 - pause, 1 - not a pause.

Return type:

np.ndarray

put(signal: MultiChannelSignal) → SignalAugmentation¶

Updates the multichannel sound data and sample rate of the MultiChannelSignal instance.

Parameters:: signal (MultiChannelSignal) – source of multichannel sound data.
Returns:: The updated SignalAugmentation instance.
Return type:: self (SignalAugmentation)

rdac(source_path: str) → list[MultiChannelSignal]¶

Reads data from a file at the specified path and updates the sample rate and data attributes and applies chains if they exist in object.

Parameters:: source_path (str) – Path to the file containing the data.
Returns:: The updated SignalAugmentation data field.
Return type:: list[MultiChannelSignal]

read_file_apply_chains(source_path: str) → list[MultiChannelSignal]¶

Reads data from a file at the specified path and updates the sample rate and data attributes and applies chains if they exist in object.

Parameters:: source_path (str) – Path to the file containing the data.
Returns:: The updated SignalAugmentation data field.
Return type:: list[MultiChannelSignal]