1mlib_SignalLPCCovariance_F3m2e(d3iMaLLIiBb)Library Fmulnicbt_iSoingsnalLPCCovariance_F32(3MLIB)
2
3
4

NAME

6       mlib_SignalLPCCovariance_F32  -  perform  linear predictive coding with
7       covariance method
8

SYNOPSIS

10       cc [ flag... ] file... -lmlib [ library... ]
11       #include <mlib.h>
12
13       mlib_status mlib_SignalLPCCovariance_F32(mlib_f32 *coeff,
14            const mlib_f32 *signal, void *state);
15
16

DESCRIPTION

18       The mlib_SignalLPCCovariance_F32() function performs linear  predictive
19       coding with covariance method.
20
21
22       In  linear  predictive coding (LPC) model, each speech sample is repre‐
23       sented as a linear combination of the past M samples.
24
25                      M
26              s(n) = SUM a(i) * s(n-i) + G * u(n)
27                     i=1
28
29
30
31       where s(*) is the speech signal, u(*) is the excitation signal,  and  G
32       is  the gain constants, M is the order of the linear prediction filter.
33       Given s(*), the goal is to find a set of coefficient  a(*)  that  mini‐
34       mizes the prediction error e(*).
35
36                             M
37              e(n) = s(n) - SUM a(i) * s(n-i)
38                            i=1
39
40
41
42       In  covariance method, the coefficients can be obtained by solving fol‐
43       lowing set of linear equations.
44
45               M
46              SUM a(i) * c(i,k) = c(0,k), k=1,...,M
47              i=1
48
49
50
51       where
52
53                       N-k-1
54              c(i,k) =  SUM s(j) * s(j+k-i)
55                        j=0
56
57
58
59       are the covariance coefficients of s(*), N is the length of  the  input
60       speech vector.
61
62
63       Note  that the covariance matrix R is a symmetric matrix, and the equa‐
64       tions can be solved efficiently with Cholesky decomposition method.
65
66
67       See Fundamentals of Speech Recognition by Lawrence Rabiner  and  Biing-
68       Hwang Juang, Prentice Hall, 1993.
69

PARAMETERS

71       The function takes the following arguments:
72
73       coeff     The linear prediction coefficients.
74
75
76       signal    The input signal vector.
77
78
79       state     Pointer to the internal state structure.
80
81

RETURN VALUES

83       The  function  returns MLIB_SUCCESS if successful. Otherwise it returns
84       MLIB_FAILURE.
85

ATTRIBUTES

87       See attributes(5) for descriptions of the following attributes:
88
89
90
91
92       ┌─────────────────────────────┬─────────────────────────────┐
93       │      ATTRIBUTE TYPE         │      ATTRIBUTE VALUE        │
94       ├─────────────────────────────┼─────────────────────────────┤
95       │Interface Stability          │Committed                    │
96       ├─────────────────────────────┼─────────────────────────────┤
97       │MT-Level                     │MT-Safe                      │
98       └─────────────────────────────┴─────────────────────────────┘
99

SEE ALSO

101       mlib_SignalLPCCovarianceInit_F32(3MLIB),      mlib_SignalLPCCovariance‐
102       Free_F32(3MLIB), attributes(5)
103
104
105
106SunOS 5.11                        2 Mar 2007mlib_SignalLPCCovariance_F32(3MLIB)