iulaw#

class diffsptk.MuLawExpansion(abs_max: float = 1, mu: int = 255)[source]#

See this page for details.

Parameters:

abs_maxfloat > 0: The absolute maximum value of the original input waveform.
muint >= 1: The compression factor, \(\mu\).

forward(y: Tensor) → Tensor[source]#

Expand the waveform using the \(\mu\)-law algorithm.

Parameters:

yTensor [shape=(…,)]: The input compressed waveform.

Returns:

outTensor [shape=(…,)]: The expanded waveform.

Examples

>>> x = diffsptk.ramp(4)
>>> ulaw = diffsptk.MuLawCompression(4)
>>> iulaw = diffsptk.MuLawExpansion(4)
>>> x2 = iulaw(ulaw(x))
>>> x2
tensor([0.0000, 1.0000, 2.0000, 3.0000, 4.0000])

diffsptk.functional.iulaw(y: Tensor, abs_max: float = 1, mu: int = 255) → Tensor[source]#

Expand the waveform using the \(\mu\)-law algorithm.

Parameters:

yTensor [shape=(…,)]: The compressed waveform.
abs_maxfloat > 0: The absolute maximum value of the original input waveform.
muint >= 1: The compression factor, \(\mu\).

Returns:

outTensor [shape=(…,)]: The expanded waveform.