wavaugmentate 0.2.1

The package wavaugmentate makes audio signal augmentation conversions.

Wavaugmentate 0.2.1

Multichannel Audio Signal Augmentation Module

The module makes audio signal augmentation conversions. It provides the Mcs, Aug classes and wavaug-cli console utility.

• Mcs is a multi-channel audio.
• Aug helps to perform augmentation of multi-channel audio signals for AI models learning purpose.

Installation

pip install wavaugmentate

Input Data

WAV-file or NumPy array.

Array shape: (num_channels, num_samples).

Output Data

Same types as in section Input_data.

Augmentation Methods

1. Amplitude (volume change, inversion).
2. Time shift.
3. Echo.
4. Adding noise.
5. Time stretching. (not implemented)
6. Tempo change. (not implemented)
7. Pitch shift. (not implemented)
8. Adding silence.
9. Frequency masking. (not implemented)
10. Time masking. (not implemented)
11. Combinations of methods.

Additional Functionality

1. Generation multichannel tonal signals of desired frequency, amplitude, durance.
2. Generation multichannel speech-like signals of desired formants frequency, amplitude, durance.

Interfaces

Signal augmentation can be applied by two ways:

1. As python module Mcs, Aug classes methods.
2. As console application wavaugmentate with CLI interface options.

Python Module

Example 1 (procedural approach):

from wavaugmentate.mcs import MultiChannelSignal as Mcs
from wavaugmentate.aug import AudioAugmentation as Aug


# File name of original sound.
file_name = "./outputwav/sound.wav"

# Create Mcs-object.
mcs = Mcs()

# Read WAV-file to Mcs-object.
mcs.read(file_name)

# Change quantity of channels to 7.
mcs.split(7)

# Create augmentation object.
aug = Aug(mcs)

# Apply delays.
# Corresponds to channels quantity.
delay_list = [0, 150, 200, 250, 300, 350, 400]
aug.delay_ctrl(delay_list)

# Apply amplitude changes.
# Corresponds to channels quantity.
amplitude_list = [1, 0.17, 0.2, 0.23, 0.3, 0.37, 0.4]
aug.amplitude_ctrl(amplitude_list)

# Augmentation result saving by single file, containing 7 channels.
aug.get().write(sound_aug_file_path)

# Augmentation result saving to 7 files, each 1 by channel.
# ./outputwav/sound_augmented_1.wav
# ./outputwav/sound_augmented_2.wav and so on.
aug.get().write_by_channel(sound_aug_file_path)

Original signal is shown on picture:

Output signal with augmented data (channel 1 contains original signal without changes):

The same code as chain of operations, Example 2:

from wavaugmentate.mcs import MultiChannel as Mcs
from wavaugmentate.aug import AudioAugmentation as Aug

# File name of original sound.
file_name = "./outputwav/sound.wav"

delay_list = [0, 150, 200, 250, 300, 350, 400]
amplitude_list = [1, 0.17, 0.2, 0.23, 0.3, 0.37, 0.4]

# Apply all transformations of Example 1 in chain.
ao_obj = Aug(Mcs().rd(file_name))
ao_obj.splt(7).dly(delay_list).amp(amplitude_list).get().wr(
"sound_augmented_by_chain.wav"
)

# Augmentation result saving to 7 files, each 1 by channel.
ao_obj.get().wrbc("sound_augmented_by_chain.wav")
 

CLI

use for help:

wavaug-cli -h

command line interface provides the same functionality.

Example 3 (procedural approach):

wavaug-cli -i ./test_sounds/test_sound_1.wav -o ./outputwav/out.wav -d "100, 200, 300, 400"
wavaug-cli -i ./outputwav/out.wav -o ./outputwav/out.wav -a "0.1, 0.2, 0.3, 0.4"

Example 4 (OOP approach):

wavaug-cli -c 'rd("./test_sounds/test_sound_1.wav").dly([100, 200, 300, 400]).amp([0.1, 0.2, 0.3, 0.4]).wr("./outputwav/sound_delayed.wav")'

How To

Single file to several augmented

Amplitudes and delays will be augmented by code shown in example 5.

Example 5 (single file augmentation):

from wavaugmentate.mcs import MultiChannel as Mcs
from wavaugmentate.aug import AudioAugmentation as Aug

file_name = "./outputwav/sound.wav"
mcs = Mcs()
mcs.rd(file_name)  # Read original file with single channel.
file_name_head = "sound_augmented"

# Suppose we need 15 augmented files.
aug_count = 15
for i in range(aug_count):
   signal = Aug(mcs.copy())
   # Apply random amplitude [0.3..1.7) and delay [70..130)
   # microseconds changes to each copy of original signal.
   signal.amp([1], [0.7]).dly([100], [30])
   name = file_name_head + f"_{i + 1}.wav"
   signal.get().write(name)    

Unit Tests

Just run:

export  PYTHONPATH='./src/wavaugmentate'
python3 -m pytest

Test coverage:

---------- coverage: platform linux, python 3.11.4-final-0 -----------
Name                       Stmts   Miss  Cover
----------------------------------------------
common_test_functions.py      15      0   100%
test_mcs_class.py            385      0   100%
test_wavaugmentate.py        293      0   100%
wavaugmentate.py             507     38    93%
----------------------------------------------
TOTAL                       1200     38    97%

Reference

MCS - multi channel signal, it is NumPy array with shape (M_channels, N_samples).

#	Mcs class method	CLI option	Method alias	Description
1	read(path)	-c 'rd(path)'	rd	Read MCS from WAV-file.
2	write(path)	-c 'wr(path)'	wr	Save MCS to WAV-file.
3	file_info(path)	--info	info	Returns WAV-file info.
4	-	-i path	-	Input WAV-file path.
5	-	-o path	-	Output WAV-file path.
6	amplitude_ctrl([c1,c2..cm])	-a "c1,c2..Cm"	amp	Change amplitudes of channels.
7	delay_ctrl([t1,t2..tm])	-d "t1,t2..tm"	dly	Add delays to channels.
8	echo _ctrl([t1,t2..tm],[c1,c2..cm])	-d "t1,t2..tm / c1,c2..Cm"	echo	Add echo to channels.
9	noise_ctrl([c1,c2..cm])	-n "c1,c2..Cm"	ns	Add normal noise to channels.
10	copy	-	cpy	Makes copy of MCS.
11	generate([f1,f2,f3..fm],duration,fs)	-	gen	Creates MCS and generates sine signal for each channel.
12	merge()	-	mrg	Merges all channels to single and returns mono MCS.
13	pause_detect(relative_level)	-	pdt	Searches pauses by selected levels. Returns array-mask.
14	pause_set(pause_map,pause_sz)	-	-	Set pauses length to selected values. Returns updated MCS.
15	rms()	-	rms	Returns list of RMS calculated for object channels.
16	side_by_side(mcs)	-	sbs	Appends channels from mcs data as new channels.
17	split(m_channels)	-	splt	Splits single channel to m_channels copies.
18	sum(mcs2)	-	sum	Adds mcs2 data channels values to object channels data sample by sample.
19	write_by_channel(path)	-	wrbc	Save MCS object channels to separate WAV-files.

Documentation

Documentation on the Read the Docs

For local documentation make clone of repository and look html-version of documentation (docs/_build/html/index.html): html-documentation

Rebuild Documentation

cd docs
make html

Build Package

Install builder:

python3 -m pip install --upgrade build

build package:

python3 -m build