ignorm#

class diffsptk.GeneralizedCepstrumInverseGainNormalization(cep_order: int, gamma: float = 0, c: int | None = None)[source]#

See this page for details.

Parameters:

cep_orderint >= 0: The order of the cepstrum, \(M\).
gammafloat in [-1, 1]: The gamma parameter, \(\gamma\).
cint >= 1 or None: The number of filter stages.

References

[1]

T. Kobayashi et al., “Spectral analysis using generalized cepstrum,” IEEE Transactions on Acoustics, Speech, and Signal Processing, vol. 32, no. 5, pp. 1087-1089, 1984.

forward(y: Tensor) → Tensor[source]#

Perform cepstrum inverse gain normalization.

Parameters:

yTensor [shape=(…, M+1)]: The normalized generalized cepstrum.

Returns:

xTensor [shape=(…, M+1)]: The generalized cepstrum.

Examples

>>> x = diffsptk.ramp(1, 4)
>>> gnorm = diffsptk.GeneralizedCepstrumGainNormalization(3, c=2)
>>> ignorm = diffsptk.GeneralizedCepstrumInverseGainNormalization(3, c=2)
>>> x2 = ignorm(gnorm(x))
>>> x2
tensor([1., 2., 3., 4.])

diffsptk.functional.ignorm(y: Tensor, gamma: float = 0, c: int | None = None) → Tensor[source]#

Perform cepstrum inverse gain normalization.

Parameters:

yTensor [shape=(…, M+1)]: The normalized generalized cepstrum.
gammafloat in [-1, 1]: The gamma parameter, \(\gamma\).
cint >= 1 or None: The number of filter stages.

Returns:

outTensor [shape=(…, M+1)]: The generalized cepstrum.