snr

Functions

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

snr [ option ] file1 [ infile ]

• -l int

  • frame length $(1\le L)$

• -o int

  • output type $(0\le O\le 2)$

    • 0 SNR

    • 1 segmental SNR

    • 2 segmental SNR per frame

• file1 str

  • double-type signal sequence, $x_1$

• infile str

  • double-type signal+noise sequence, $x_2$

• stdout

  • double-type SNR

The inputs of this commands are two signals:

$$\begin{array}{r}\begin{array}{cccc}{x}_1(0),& x_1(1),& \dots ,& x_1(T-1),\\ x_2(0),& x_2(1),& \dots ,& x_2(T-1).\end{array}\end{array}$$

If $O=2$, segmental SNR is calculated and output at every frame:

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

where $N=\lfloor (T-1)/L\rfloor$ and

$$y(n)=10{\log }_{10}\frac{{\displaystyle \sum _{l=0}^{L-1}\{x_1(nL+l){\}}^2}}{{\displaystyle \sum _{l=0}^{L-1}\{x_2(nL+l)-x_1(nL+l){\}}^2}}.$$

If $O=1$, the output is averaged segmental SNR:

$$y=\frac{1}{N}\sum _{n=0}^{N-1}y(n).$$

If $O=0$, the output is a standard SNR:

$$y=10{\log }_{10}\frac{{\displaystyle \sum _{t=0}^{T-1}\{x_1(t){\}}^2}}{{\displaystyle \sum _{t=0}^{T-1}\{x_2(t)-x_1(t){\}}^2}}.$$

Parameters:

• argc – [in] Number of arguments.

• argv – [in] Argument vector.

Returns:

0 on success, 1 on failure.

See also

rmse