1mlib_SignalLPCAutoCorrel_S1m6e(d3iMaLLIiBb)Library Fmulnicbt_iSoingsnalLPCAutoCorrel_S16(3MLIB)
2
3
4

NAME

6       mlib_SignalLPCAutoCorrel_S16,  mlib_SignalLPCAutoCorrel_S16_Adp  - per‐
7       form linear predictive coding with autocorrelation method
8

SYNOPSIS

10       cc [ flag... ] file... -lmlib [ library... ]
11       #include <mlib.h>
12
13       mlib_status mlib_SignalLPCAutoCorrel_S16(mlib_s16 *coeff,
14            mlib_s32 cscale, const mlib_s16 *signal, void *state);
15
16
17       mlib_status mlib_SignalLPCAutoCorrel_S16_Adp(mlib_s16 *coeff,
18            mlib_s32 *cscale, const mlib_s16 *signal, void *state);
19
20

DESCRIPTION

22       Each function performs linear predictive  coding  with  autocorrelation
23       method.
24
25
26       In  linear  predictive coding (LPC) model, each speech sample is repre‐
27       sented as a linear combination of the past M samples.
28
29                      M
30              s(n) = SUM a(i) * s(n-i) + G * u(n)
31                     i=1
32
33
34
35       where s(*) is the speech signal, u(*) is the excitation signal,  and  G
36       is  the gain constants, M is the order of the linear prediction filter.
37       Given s(*), the goal is to find a set of coefficient  a(*)  that  mini‐
38       mizes the prediction error e(*).
39
40                             M
41              e(n) = s(n) - SUM a(i) * s(n-i)
42                            i=1
43
44
45
46       In  autocorrelation method, the coefficients can be obtained by solving
47       following set of linear equations.
48
49               M
50              SUM a(i) * r(|i-k|) = r(k), k=1,...,M
51              i=1
52
53
54
55       where
56
57                    N-k-1
58              r(k) = SUM s(j) * s(j+k)
59                     j=0
60
61
62
63       are the autocorrelation coefficients of s(*), N is the  length  of  the
64       input speech vector. r(0) is the energy of the speech signal.
65
66
67       Note  that the autocorrelation matrix R is a Toeplitz matrix (symmetric
68       with all diagonal elements equal), and  the  equations  can  be  solved
69       efficiently with Levinson-Durbin algorithm.
70
71
72       See  Fundamentals  of Speech Recognition by Lawrence Rabiner and Biing-
73       Hwang Juang, Prentice Hall, 1993.
74
75
76       Note for functions with adaptive scaling (with _Adp postfix), the scal‐
77       ing  factor  of  the output data will be calculated based on the actual
78       data; for functions with non-adaptive scaling (without  _Adp  postfix),
79       the  user  supplied  scaling factor will be used and the output will be
80       saturated if necessary.
81

PARAMETERS

83       Each function takes the following arguments:
84
85       coeff     The linear prediction coefficients.
86
87
88       cscale    The scaling factor of  the  linear  prediction  coefficients,
89                 where actual_data = output_data * 2**(-scaling_factor).
90
91
92       signal    The input signal vector with samples in Q15 format.
93
94
95       state     Pointer to the internal state structure.
96
97

RETURN VALUES

99       Each  function returns MLIB_SUCCESS if successful. Otherwise it returns
100       MLIB_FAILURE.
101

ATTRIBUTES

103       See attributes(5) for descriptions of the following attributes:
104
105
106
107
108       ┌─────────────────────────────┬─────────────────────────────┐
109       │      ATTRIBUTE TYPE         │      ATTRIBUTE VALUE        │
110       ├─────────────────────────────┼─────────────────────────────┤
111       │Interface Stability          │Committed                    │
112       ├─────────────────────────────┼─────────────────────────────┤
113       │MT-Level                     │MT-Safe                      │
114       └─────────────────────────────┴─────────────────────────────┘
115

SEE ALSO

117       mlib_SignalLPCAutoCorrelInit_S16(3MLIB), mlib_SignalLPCAutoCorrelGetEn‐
118       ergy_S16(3MLIB),          mlib_SignalLPCAutoCorrelGetPARCOR_S16(3MLIB),
119       mlib_SignalLPCAutoCorrelFree_S16(3MLIB), attributes(5)
120
121
122
123SunOS 5.11                        2 Mar 2007mlib_SignalLPCAutoCorrel_S16(3MLIB)
Impressum