mlib_SignalLPC2Cepstral_S16_Adp

- convert linear prediction coefficients to cepstral coefficients

Synopsis

cc [ flag... ] file... -lmlib [ library... ]
#include <mlib.h>

mlib_status mlib_SignalLPC2Cepstral_S16_Adp(mlib_s16 *cepst,
     mlib_s32 *cscale, const mlib_s16 *lpc, mlib_s32 lscale,
     mlib_s16 gain, mlib_s32 gscale, mlib_s32 length,
     mlib_s32 order);

Description

The mlib_SignalLPC2Cepstral_S16_Adp() function converts linear prediction coefficients to cepstral coefficients. The scaling factor of the output data, cscale, will be calculated based on the actual data.

The cepstral coefficients are the coefficients of the Fourier transform representation of the log magnitude spectrum.

The LPC cepstral coefficients can be derived recursively from the LPC coefficients as following.

c(0) = log(G)

              m-1  k
c(m) = a(m) + SUM --- * c(k) * a(m-k), 1 ≤ m ≤ M
              k=1  m

       m-1  k
c(m) = SUM --- * c(k) * a(m-k), m > M
       k=1  m

See Fundamentals of Speech Recognition by Lawrence Rabiner and Biing-Hwang Juang, Prentice Hall, 1993.

Parameters

The function takes the following arguments:

cepst

The cepstral coefficients.

cscale

The scaling factor of the cepstral coefficients, where actual_data = output_data * 2**(-scaling_factor).

lpc

The linear prediction coefficients.

lscale

The scaling factor of the linear prediction coefficients, where actual_data = input_data * 2**(-scaling_factor).

gain

The gain of the LPC model.

gscale

The scaling factor of the gain of the LPC model, where actual_data = input_data * 2**(-scaling_factor).

length

The length of the cepstral coefficients.

order

The order of the linear prediction filter.

Return Values

The function returns MLIB_SUCCESS if successful. Otherwise it returns MLIB_FAILURE.

Attributes

See attributes(5) for descriptions of the following attributes:

See Also

mlib_SignalLPC2Cepstral_S16(3MLIB), mlib_SignalLPC2Cepstral_S16_Adp(3MLIB), mlib_SignalLPC2Cepstral_F32(3MLIB), attributes(5)