lpccheck#

class diffsptk.LinearPredictiveCoefficientsStabilityCheck(lpc_order, margin=1e-16, warn_type='warn')[source]#

See this page for details.

Parameters:
lpc_orderint >= 0 [scalar]

Order of LPC, \(M\).

margin[0 < float < 1]

Margin.

warn_type[‘ignore’, ‘warn’, ‘exit’]

Behavior for unstable LPC.

forward(a1)[source]#

Check stability of LPC.

Parameters:
a1Tensor [shape=(…, M+1)]

LPC coefficients.

Returns:
a2Tensor [shape=(…, M+1)]

Modified LPC coefficients.

Examples

>>> x = diffsptk.nrand(4)
tensor([-0.9966, -0.2970, -0.2173,  0.0594,  0.5831])
>>> lpc = diffsptk.LPC(3, 5)
>>> a = lpc(x)
>>> a
tensor([ 1.1528, -0.2613, -0.0274,  0.1778])
>>> lpccheck = diffsptk.LinearPredictiveCoefficientsStabilityCheck(3)
>>> a2 = lpccheck(a)
tensor([ 1.1528, -0.2613, -0.0274,  0.1778])

See also

lpc lpc2par par2lpc