julius/search.h File Reference

Strucures for handling hypotheses on the 2nd pass. More...

This graph shows which files directly or indirectly include this file:

Go to the source code of this file.

Data Structures
struct	__nextword__
struct	__node__
Typedefs
typedef __nextword__	NEXTWORD
typedef __node__	NODE

---

Detailed Description

Strucures for handling hypotheses on the 2nd pass.

Author:

Akinobu Lee

Date:

Wed Sep 07 07:40:11 2005

This file includes definitions for handling hypothesis used on the 2nd pass stack decoding. Partial sentence hypotheses are stored in NODE structure, with its various information about total scores, viterbi scores, language scores, confidence scores, estimated end frame, and so on. WordGraph express a word in graph, generated through the 2nd pass. NEXTWORD is used to hold next word information at hypothesis expantion stage. POPNODE will be used when Visualization is enabled to store the search trail.

Revision

1.4

Definition in file search.h.

---

Typedef Documentation

typedef struct __nextword__ NEXTWORD

Next word candidate in the 2nd pass. This will be used to hold word candidates that can be connected to a given hypothesis.

typedef struct __node__ NODE

Sentence hypothesis at 2nd pass

---