cutcutcodec.core.generation.audio.noise¶

Generate an audio noise signal.

Classes

GeneratorAudioNoise([seed, layout])

Generate a pure noise audio signal.

Functions

md5(string)

Vectorised version of the md5 algorithm.

Details

class cutcutcodec.core.generation.audio.noise.GeneratorAudioNoise(seed: Real | None = None, layout: Layout | str | Integral | None = 'stereo')[source]

Generate a pure noise audio signal.

Attributes¶

layoutcutcutcodec.core.classes.layout.Layout: The signification of each channels (readonly).

Examples¶

>>> from fractions import Fraction
>>> from cutcutcodec.core.generation.audio.noise import GeneratorAudioNoise
>>> (stream,) = GeneratorAudioNoise(0).out_streams
>>> stream.snapshot(Fraction(2, 48000), 48000, 5).numpy(force=True)
array([[-0.25397146, -0.1199106 , -0.12052107, -0.16986334, -0.50950706],
       [ 0.17409873,  0.42185044, -0.7231959 ,  0.39764726, -0.25968206]],
      dtype=float32)
>>> stream.snapshot(0, 24000, 5).numpy(force=True)
array([[ 0.44649088, -0.25397146, -0.12052107, -0.50950706,  0.5112531 ],
       [-0.8036704 ,  0.17409873, -0.7231959 , -0.25968206, -0.7944578 ]],
      dtype=float32)
>>> frame = stream.snapshot(0, 48000, 48000*60)  # test uniform
>>> abs(round(frame.mean().item(), 3))  # theory 0
0.0
>>> round(frame.var().item(), 3)  # theory 1/3
0.333
>>>

Initialise and create the class.

Parameters¶

seednumbers.Real, optional: Transmitted to cutcutcodec.core.interfaces.seedable.Seedable.
layout: cutcutcodec.core.classes.layout.Layout or str or int, optional: The audio layout to associate to each equation, let see cutcutcodec.core.classes.layout.Layout for more details. By default, the layout is stereo, two channels.

cutcutcodec.core.generation.audio.noise.md5(string: ndarray) → ndarray[source]

Vectorised version of the md5 algorithm.

Parameters¶

stringnp.ndarray: The values to hash, of shape (16, n) and dtype uint32.

Returns¶

hashtuple[np.ndarray, np.ndarray, np.ndarray, np.ndarray]: The hashed value of the input string, of shape (n,). Fours differents 32 bits values ares yields, total 128 bits.

Examples¶

>>> import numpy as np
>>> from cutcutcodec.core.generation.audio.noise import md5
>>> string = np.zeros((16, 1), dtype=np.uint32)
>>> np.vstack(md5(string))
array([[1543640769],
       [2049272632],
       [1899532917],
       [2915357108]], dtype=uint32)
>>>