libsent/src/phmm/gprune_heu.c File Reference

Calculate probability of a set of Gaussian densities by Gaussian pruning: heuristic algorithm. More...

#include <sent/stddefs.h>
#include <sent/htk_hmm.h>
#include <sent/htk_param.h>
#include <sent/hmm.h>
#include <sent/hmm_calc.h>

Go to the source code of this file.

Functions

static void init_backmax (HMMWork *wrk)
 Clear backmax information.
static void make_backmax (HMMWork *wrk)
 Build backmax information for each frame, by summing up current maximum values of each dimensions.
static LOGPROB compute_g_heu_updating (HMMWork *wrk, HTK_HMM_Dens *binfo)
 Calculate probability with maximum value update.
static LOGPROB compute_g_heu_pruning (HMMWork *wrk, HTK_HMM_Dens *binfo, LOGPROB thres)
 Calculate probability with pruning.
boolean gprune_heu_init (HMMWork *wrk)
 Initialize and setup work area for Gaussian pruning by heuristic algorithm.
void gprune_heu_free (HMMWork *wrk)
 Free gprune_heu related work area.
void gprune_heu (HMMWork *wrk, HTK_HMM_Dens **g, int gnum, int *last_id)
 Compute a set of Gaussians with heuristic pruning.


Detailed Description

Calculate probability of a set of Gaussian densities by Gaussian pruning: heuristic algorithm.

gprune_heu() is one of the functions to compute output probability of a set of Gaussian densities. This function does heuristic pruning, trying to compute only the best ones for faster computation. If a tied-mixture HMM model with GPRUNE_DEFAULT_HEURISTIC defined in Julius, or explicitly specified by "-gprune heuristic" option, this function will be used.

The effect of heuristic pruning is practically a middle of others.

gprune_heu() will be used by calling outprob_init() to set its pointer to the global variable compute_gaussset. Then it will be called from calc_tied_mix() or calc_mix().

Author:
Akinobu LEE
Date:
Thu Feb 17 05:44:52 2005
Revision
1.1.1.1

Definition in file gprune_heu.c.


Function Documentation

static void init_backmax ( HMMWork wrk  )  [static]

Clear backmax information.

Parameters:
wrk [i/o] HMM computation work area

Definition at line 112 of file gprune_heu.c.

Referenced by gprune_heu().

static void make_backmax ( HMMWork wrk  )  [static]

Build backmax information for each frame, by summing up current maximum values of each dimensions.

Parameters:
wrk [i/o] HMM computation work area

Definition at line 133 of file gprune_heu.c.

Referenced by gprune_heu().

static LOGPROB compute_g_heu_updating ( HMMWork wrk,
HTK_HMM_Dens binfo 
) [static]

Calculate probability with maximum value update.

Calculate probability of a Gaussian toward OP_vec, while storing the maximum values of each dimension to backmax. for future pruning. The pruning itself is not performed here. This function will be used to compute the first N Gaussians.

Parameters:
wrk [i/o] HMM computation work area
binfo [in] Gaussian density
Returns:
the output log probability.

Definition at line 161 of file gprune_heu.c.

Referenced by gprune_heu().

static LOGPROB compute_g_heu_pruning ( HMMWork wrk,
HTK_HMM_Dens binfo,
LOGPROB  thres 
) [static]

Calculate probability with pruning.

Calculate probability of a Gaussian toward OP_vec, performing pruning using threshold and the backmax information that has been set by compute_g_heu_updating() and make_backmax().

Parameters:
wrk [i/o] HMM computation work area
binfo [in] Gaussian density
thres [in] threshold
Returns:
the output log probability.

Definition at line 200 of file gprune_heu.c.

Referenced by gprune_heu().

boolean gprune_heu_init ( HMMWork wrk  ) 

Initialize and setup work area for Gaussian pruning by heuristic algorithm.

Parameters:
wrk [i/o] HMM computation work area
Returns:
TRUE on success, FALSE on failure.

Definition at line 237 of file gprune_heu.c.

Referenced by outprob_init().

void gprune_heu_free ( HMMWork wrk  ) 

Free gprune_heu related work area.

Parameters:
wrk [i/o] HMM computation work area

Definition at line 259 of file gprune_heu.c.

Referenced by outprob_init().

void gprune_heu ( HMMWork wrk,
HTK_HMM_Dens **  g,
int  gnum,
int *  last_id 
)

Compute a set of Gaussians with heuristic pruning.

If the N-best mixtures in the previous frame is specified in last_id, They are first computed to get the maximum value for each dimension. After that, the rest of the Gaussians will be computed using the maximum values as heuristics of uncomputed dimensions to drop unpromising Gaussians from computation at early stage of likelihood computation. If the last_id is not specified (typically at the first frame of the input), a safe pruning as same as one in gprune_safe.c will be applied.

The calculated scores will be stored to OP_calced_score, with its corresponding mixture id to OP_calced_id. These are done by calling cache_push(). The number of calculated mixtures is also stored in OP_calced_num.

This can be called from calc_tied_mix() or calc_mix().

Parameters:
wrk [i/o] HMM computation work area
g [in] set of Gaussian densities to compute the output probability
gnum [in] length of above
last_id [in] ID list of N-best mixture in previous input frame, or NULL if not exist

Definition at line 293 of file gprune_heu.c.

Referenced by outprob_init().


Generated on Tue Dec 18 16:01:40 2007 for Julius by  doxygen 1.5.4