csm2acr#

class diffsptk.CompositeSinusoidalModelCoefficientsToAutocorrelation(acr_order: int, device: device | None = None, dtype: dtype | None = None)[source]#

See this page for details.

Parameters:

acr_orderint >= 0: The order of the autocorrelation, \(M\).
devicetorch.device or None: The device of this module.
dtypetorch.dtype or None: The data type of this module.

References

[1]

S. Sagayama et al., “Duality theory of composite sinusoidal modeling and linear prediction,” Proceedings of ICASSP, pp. 1261-1264, 1986.

forward(c: Tensor) → Tensor[source]#

Convert CSM coefficients to autocorrelation.

Parameters:

cTensor [shape=(…, M+1)]: The CSM coefficients.

Returns:

outTensor [shape=(…, M+1)]: The autocorrelation.

Examples

>>> import diffsptk
>>> acorr = diffsptk.Autocorrelation(5, 3)
>>> acr2csm = diffsptk.AutocorrelationToCompositeSinusoidalModelCoefficients(3)
>>> csm2acr = diffsptk.CompositeSinusoidalModelCoefficientsToAutocorrelation(3)
>>> x = diffsptk.ramp(4)
>>> r = acorr(x)
>>> r
tensor([30.0000, 20.0000, 11.0000,  4.0000])
>>> r2 = csm2acr(acr2csm(r))
>>> r2
tensor([30.0000, 20.0000, 11.0000,  4.0000])

diffsptk.functional.csm2acr(c: Tensor) → Tensor[source]#

Convert CSM coefficients to autocorrelation.

Parameters:

cTensor [shape=(…, M+1)]: The CSM coefficients.

Returns:

outTensor [shape=(…, M+1)]: The autocorrelation.