mlsacheck#

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

See this page for details.

Parameters:

cep_orderint >= 0: Order of mel-cepstrum, \(M\).
alphafloat in (-1, 1): Frequency warping factor, \(\alpha\).
pade_orderint in [4, 7]: Order of Pade approximation.
strictbool: If True, keep maximum log approximation error rather than MLSA filter stability.
thresholdfloat > 0 or None: Threshold value. If not given, automatically computed.
fastbool: Enable fast mode (do not use FFT).
n_fftint > M: Number of FFT bins, \(L\). Used only in non-fast mode.
warn_type[‘ignore’, ‘warn’, ‘exit’]: Warning type.
mod_type[‘clip’, ‘scale’]: Modification type.

forward(c)[source]#

Check stability of MLSA filter.

Parameters:

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

Returns:

outTensor [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])

diffsptk.functional.mlsacheck(c, *, alpha=0, pade_order=4, strict=True, threshold=None, fast=True, n_fft=512, warn_type='warn', mod_type='scale')[source]#

Check stability of MLSA filter.

Parameters:

cTensor [shape=(…, M+1)]: Mel-cepstrum.
alphafloat in (-1, 1): Frequency warping factor, \(\alpha\).
pade_orderint in [4, 7]: Order of Pade approximation.
strictbool: If True, keep maximum log approximation error rather than MLSA filter stability.
thresholdfloat > 0 or None: Threshold value. If not given, automatically computed.
fastbool: Enable fast mode.
n_fftint > M: Number of FFT bins, \(L\). Used only in non-fast mode.
warn_type[‘ignore’, ‘warn’, ‘exit’]: Warning type.
mod_type[‘clip’, ‘scale’]: Modification type.

Returns:

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