mlsacheck#

class diffsptk.MLSADigitalFilterStabilityCheck(cep_order, alpha=0, fft_length=256, pade_order=4, strict=True, fast=True, threshold=None, mod_type='scale', warn_type='warn')[source]#

See this page for details.

Parameters:

cep_orderint >= 0 [scalar]: Order of mel-cepstrum, \(M\).
alphafloat [-1 < alpha < 1]: Frequency warping factor, \(\alpha\).
fft_lengthint > M [scalar]: Number of FFT bins, \(L\).
pade_order[4 <= int <= 7].: Order of Pade approximation.
strictbool [scalar]: If True, keep maximum log approximation error rather than MLSA filter stability.
fastbool [scalar]: Fast mode.
thresholdfloat > 0 [scalar]: Threshold value. If not given, automatically computed.
mod_type[‘clip’, ‘scale’]: Modification type.
warn_type[‘ignore’, ‘warn’, ‘exit’]: Behavior for unstable MLSA.

forward(c1)[source]#

Check stability of MLSA filter.

Parameters:

c1Tensor [shape=(…, M+1)]: Mel-cepstrum.

Returns:

c2Tensor [shape=(…, M+1)]: Modified mel-cepstrum.

Examples

>>> c1 = diffsptk.nrand(4, stdv=10)
>>> c1
tensor([ 1.8963,  7.6629,  4.4804,  8.0669, -1.2768])
>>> mlsacheck = diffsptk.MLSADigitalFilterStabilityCheck(4, warn_type="ignore")
>>> c2 = mlsacheck(c1)
>>> c2
tensor([ 1.3336,  1.7537,  1.0254,  1.8462, -0.2922])