Sentence Struct Reference

Output result structure. More...

#include <recog.h>

Collaboration diagram for Sentence:

Collaboration graph
[legend]

Data Fields

WORD_ID word [MAXSEQNUM]
 Sequence of word ID.
int word_num
 Number of words in the sentence.
LOGPROB score
 Likelihood (LM+AM).
LOGPROB confidence [MAXSEQNUM]
 Word confidence scores.
LOGPROB score_lm
 Language model likelihood (scaled) for N-gram.
LOGPROB score_am
 Acoustic model likelihood for N-gram.
int gram_id
 The grammar ID this sentence belongs to for DFA.
struct {
   boolean   filled
 True if has data.
   int   num
 Number of units.
   short   unittype
 Unit type (one of PER_*).
   WORD_ID *   w
 word sequence by id (PER_WORD)
   HMM_Logical **   ph
 Phone sequence (PER_PHONEME, PER_STATE).
   short *   loc
 sequence of state location in a phone (PER_STATE)
   boolean *   is_iwsp
 TRUE if PER_STATE and this is the inter-word pause state at multipath mode.
   int *   begin_frame
 List of beginning frame.
   int *   end_frame
 List of ending frame.
   LOGPROB *   avgscore
 Score averaged by frames.
   LOGPROB   allscore
 Re-computed acoustic score.
align

Detailed Description

Output result structure.

Definition at line 283 of file recog.h.


The documentation for this struct was generated from the following file:
Generated on Tue Dec 18 16:02:35 2007 for Julius by  doxygen 1.5.4