lpc¶

diffsptk.LPC¶: alias of diffsptk.core.lpc.LinearPredictiveCodingAnalysis

class diffsptk.LinearPredictiveCodingAnalysis(acr_order, frame_length, out_format='K,a')[source]¶

See this page for details. This module is a simple cascade of acorr and levdur.

Parameters

acr_orderint >= 0 [scalar]: Order of autocorrelation, \(M\).
frame_lengthint > M [scalar]: Frame length, \(L\).
out_format[‘K’, ‘a’, ‘Ka’, ‘K,a’]: K is gain and a is LPC coefficients. If this is Ka, the two output tensors are concatenated and return the tensor instead of the tuple.

forward(x)[source]¶

Perform LPC analysis.

Parameters

xTensor [shape=(…, L)]: Framed waveform

Returns

aTensor or tuple[Tensor]: Gain and/or LPC coefficients.

Examples

>>> x = torch.nrand(5)
tensor([ 0.8226, -0.0284, -0.5715,  0.2127,  0.1217])
>>> lpc = diffsptk.LPC(2, 5)
>>> a = lpc(x)
>>> a
(tensor([0.8726]), tensor([0.1475, 0.5270]))