zignal

This is a python audio signal processing library.

Example usage

>>> import zignal
>>>
>>> x = zignal.Sinetone(fs=44100, f0=997, duration=0.1, gaindb=-20)
>>> print(x)
=======================================
classname        : Sinetone
sample rate      : 44100.0 [Hz]
channels         : 1
duration         : 0.100 [s]
datatype         : float64
samples per ch   : 4410
data size        : 0.034 [Mb]
has comment      : no
peak             : [ 0.1]
RMS              : [ 0.0707]
crestfactor      : [ 1.4147]
-----------------:---------------------
frequency        : 997.0 [Hz]
phase            : 0.0 [deg]
-----------------:---------------------

>>> x.fade_out(millisec=10)
>>> x.convert_to_float(targetbits=32)
>>> x.write_wav_file("sinetone.wav")
>>> x.plot()
>>> x.plot_fft()
>>>
>>> f = zignal.filters.biquads.RBJ(filtertype="peak", gaindb=-6, f0=997, Q=0.707, fs=96000)
>>> print(f)
=======================================
classname        : RBJ
sample rate      : 96000.0 [Hz]
feedforward  (B) : [ 0.96949457 -1.87369167  0.90819329]
feedback     (A) : [ 1.         -1.87369167  0.87768787]
number of zeros  : 2
number of poles  : 2
minimum phase?   : Yes
-----------------:---------------------
stable?          : Yes
type             : peak
gain             : -6.00 [dB]
f0               : 997.0 [Hz]
Q                : 0.7070

>>> f.plot_mag_phase()
>>> f.plot_pole_zero()
>>>

See the examples folder for more examples.

Requirements

This library relies on numpy, scipy, matplotlib and optionally pyaudio. It is recommended to create a virtual environment and let pip install the dependencies automatically.

python3 -m venv <name-of-virtualenv>
. <name-of-virtualenv>/bin/activate
pip install zignal

Optionally, to be able to use a soundcard, first install the python development headers and the portaudio development files. On debian/ubuntu,

sudo apt install python3-dev portaudio19-dev

then run

pip install zignal[sndcard]

which will automatically build the portaudio library and then pyaudio.

Local development

Create a python3 virtualenv and install from the local source code to make the zignal library editable. Note that the python development headers (python3-dev) and portaudio19-dev must be installed first.

python3 -m venv venv_dev
. venv_dev/bin/activate
pip install --editable .[dev]

By running make it is now possible to run isort, flake8 and also all the unit tests. They can also be executed directly from the command line, see the Makefile for the full commands to run.

Build a release

python3 -m venv venv_build
. ./venv_build/bin/activate
pip install --upgrade pip build
python3 -m build

Design goals

1. Readability over efficiency. This is a python library for development and understanding of audio signal processing.
2. The initial goal is to write the functionality in pure python, with the use of numpy, scipy and matplotlib. See rule 1. If efficiency becomes an issue a c/c++ library might be implemented but the pure python code must remain the default choice.
3. Design for non real-time processing. Functionality to do real-time processing can be added if it does not break rule 1.
4. Self documentation. The code should aim to be well documented, in the source code itself.