mcpf#

class diffsptk.MelCepstrumPostfiltering(cep_order, alpha=0, beta=0, onset=2, ir_length=128)[source]#

See this page for details.

Parameters:

cep_orderint >= 0: Order of mel-cepstrum, \(M\).
alphafloat in (-1, 1): Frequency warping factor, \(\alpha\).
betafloat: Intensity parameter, \(\beta\).
onsetint >= 0: Onset index.
ir_lengthint >= 1: Length of impulse response.

forward(mc)[source]#

Perform mel-cesptrum postfiltering.

Parameters:

mcTensor [shape=(…, M+1)]: Mel-cepstral coefficients.

Returns:

outTensor [shape=(…, M+1)]: Postfiltered mel-cepstral coefficients.

Examples

>>> X = diffsptk.nrand(4).square()
>>> X
tensor([0.2725, 2.5650, 0.3552, 0.3757, 0.1904])
>>> mcep = diffsptk.MelCepstralAnalysis(3, 8, 0.1)
>>> mcpf = diffsptk.MelCepstrumPostfiltering(3, 0.1, 0.2)
>>> mc1 = mcep(X)
>>> mc1
tensor([-0.2819,  0.3486, -0.2487, -0.3600])
>>> mc2 = mcpf(mc1)
>>> mc2
tensor([-0.3256,  0.3486, -0.2984, -0.4320])