csm2acr

Functions

int main(int argc, char *argv[])

csm2acr [ option ] [ infile ]

• -m int

  • order of autocorrelation $(1\le M)$

• infile str

  • double-type CSM parameters

• stdout

  • double-type autocorrelation

The below example converts CSM parameters into autocorrelation coefficients:

csm2acr < data.csm > data.acr

The converted autocorrelation coefficients can be reverted by

acr2csm < data.acr > data.csm

Parameters:

• argc – [in] Number of arguments.

• argv – [in] Argument vector.

Returns:

0 on success, 1 on failure.

See also

acr2csm acorr

class CompositeSinusoidalModelingToAutocorrelation

Convert CSM parameters to autocorrelation.

The input is the $2N$ CSM parameters:

$$\begin{array}{r}\begin{array}{cccc}\omega (1),& \omega (2),& \dots ,& \omega (N),\\ m(1),& m(2),& \dots ,& m(N),\end{array}\end{array}$$

where $\omega (n)$ is the CSM frequency and $m(n)$ is the CSM intensity. The output is the $(2N-1)$-th order biased sample autocorrelation coefficients:

$$\begin{array}{cccc}v(0),& v(1),& \dots ,& v(2N-1).\end{array}$$

The sample autocorrelation is given by

$$v(l)=\sum _{i=1}^{N}m(i)\cos (l\omega (i)).$$

[1] S. Sagayama and F. Itakura, “Duality theory of composite sinusoidal modeling and linear prediction,” Proc. of ICASSP 1986, pp. 1261-1264, 1986.

Public Functions

explicit CompositeSinusoidalModelingToAutocorrelation(int num_sine_wave)

Parameters:

num_sine_wave – [in] Number of sine waves, $N$.

inline int GetNumSineWaves() const

Returns:

Number of sine waves.

inline bool IsValid() const

Returns:

True if this object is valid.

bool Run(const std::vector<double> &composite_sinusoidal_modeling, std::vector<double> *autocorrelation) const

Parameters:

• composite_sinusoidal_modeling – [in] CSM parameters.

• autocorrelation – [out] Autocorrelation coefficients.

Returns:

True on success, false on failure.

bool Run(std::vector<double> *input_and_output) const

Parameters:

input_and_output – [inout] Input and output.

Returns:

True on success, false on failure.