spaudiopy.sig

Signal class. Avoid code duplications (and errors) by defining a few custom classes here.

import numpy as np
import matplotlib.pyplot as plt
plt.rcParams['axes.grid'] = True

import spaudiopy as spa

Functions

trim_audio(A, start, stop)

Trim copy of MultiSignal audio to start and stop in seconds.

Classes

`AmbiBSignal`(signals[, fs])	Signal class for first order Ambisonics B-format signals.
`HRIRs`(left, right, azi, zen, fs)	Signal class for head-related impulse responses.
`MonoSignal`(signal, fs)	Signal class for a MONO channel audio signal.
`MultiSignal`(signals[, fs])	Signal class for a MULTI channel audio signal.

class spaudiopy.sig.MonoSignal(signal, fs)[source]

Bases: object

Signal class for a MONO channel audio signal.

__init__(signal, fs)[source]

Constructor.

Parameters:

signal (array_like)
fs (int)

classmethod from_file(filename, fs=None)[source]: Alternative constructor, load signal from filename.

copy()[source]: Return an independent (deep) copy of the instance.

save(filename, subtype='FLOAT')[source]: Save to file.

trim(start, stop)[source]: Trim audio to start and stop in seconds.

apply(func, *args, **kwargs)[source]: Apply function ‘func’ to signal, arguments are forwarded.

conv(h, **kwargs)[source]: Convolve signal, kwargs are forwarded to signal.convolve.

resample(fs_new)[source]: Resample signal to new sampling rate fs_new.

play(gain=1, wait=True)[source]: Play sound signal. Adjust gain and wait until finished.

class spaudiopy.sig.MultiSignal(signals, fs=None)[source]

Bases: MonoSignal

Signal class for a MULTI channel audio signal.

__init__(signals, fs=None)[source]

Constructor.

Parameters:

signals (list of array_like)
fs (int)

classmethod from_file(filename, fs=None)[source]: Alternative constructor, load signal from filename.

get_signals()[source]: Return ndarray of signals, stacked along rows (nCH, nSmps).

trim(start, stop)[source]: Trim all channels to start and stop in seconds.

apply(func, *args, **kwargs)[source]: Apply function ‘func’ to all signals, arguments are forwarded.

conv(irs, **kwargs)[source]: Convolve signal, kwargs are forwarded to signal.convolve.

resample(fs_new)[source]: Resample signal to new sampling rate fs_new.

play(gain=1, wait=True)[source]: Play sound signal. Adjust gain and wait until finished.

copy(): Return an independent (deep) copy of the instance.

save(filename, subtype='FLOAT'): Save to file.

class spaudiopy.sig.AmbiBSignal(signals, fs=None)[source]

Bases: MultiSignal

Signal class for first order Ambisonics B-format signals.

__init__(signals, fs=None)[source]

Constructor.

Parameters:

signals (list of array_like)
fs (int)

classmethod from_file(filename, fs=None)[source]: Alternative constructor, load signal from filename.

classmethod sh_to_b(multisig)[source]

Alternative constructor, convert from sig.Multisignal.

Assumes ACN channel order.

apply(func, *args, **kwargs): Apply function ‘func’ to all signals, arguments are forwarded.

conv(irs, **kwargs): Convolve signal, kwargs are forwarded to signal.convolve.

copy(): Return an independent (deep) copy of the instance.

get_signals(): Return ndarray of signals, stacked along rows (nCH, nSmps).

play(gain=1, wait=True): Play sound signal. Adjust gain and wait until finished.

resample(fs_new): Resample signal to new sampling rate fs_new.

save(filename, subtype='FLOAT'): Save to file.

trim(start, stop): Trim all channels to start and stop in seconds.

class spaudiopy.sig.HRIRs(left, right, azi, zen, fs)[source]

Bases: object

Signal class for head-related impulse responses.

__init__(left, right, azi, zen, fs)[source]

Constructor.

Parameters:

left ((numDirs, numTaps) ndarray) – Left ear HRIRs.
right ((numDirs, numTaps) ndarray) – Right ear HRIRs.
azi ((numDirs,) array_like, in rad)
fs (int)

copy()[source]: Return an independent (deep) copy of the instance.

update_hrirs(left, right)[source]

Update and replace HRIRs in place.

Parameters:

left ((numDirs, numTaps) ndarray) – Left ear HRIRs.
right (numDirs, numTaps ndarray) – Right ear HRIRs.

Returns:

None.

nearest_hrirs(azi, zen)[source]

For a point on the sphere, select closest HRIR defined on grid.

Based on the haversine distance.

Parameters:

azi (float) – Azimuth.
zen (float) – Zenith / Colatitude.

Returns:

h_l ((n,) array_like) – h(t) closest to [phi, theta].
h_r ((n,) array_like) – h(t) closest to [phi, theta].

nearest_idx(azi, zen)[source]

Index of nearest HRIR grid point based on dot product.

Parameters:

azi (float, array_like) – Azimuth.
zen (float, array_like) – Zenith / Colatitude.

Returns:

idx (int, np.ndarray) – Index.

apply_ctf_eq(eq_taps=None, mode='full')[source]

Equalize common transfer function (CTF) of HRIRs.

Parameters:

eq_taps (array_like, optional) – FIR filter, None will calculate. The default is None.
mode (string, optional) – Forwarded to scipy.signal.convolve(). The default is ‘full’.

Returns:

None.

spaudiopy.sig.trim_audio(A, start, stop)[source]: Trim copy of MultiSignal audio to start and stop in seconds.