1    mkbingram
2
3MKBINGRAM(1)                General Commands Manual               MKBINGRAM(1)
4
5
6

NAME

8           mkbingram
9          - make binary N-gram from ARPA N-gram file
10

SYNOPSIS

12       mkbingram [-nlr forward_ngram.arpa] [-nrl backward_ngram.arpa]
13                 [-d old_bingram_file] {output_bingram_file}
14

DESCRIPTION

16       mkbingram is a tool to convert N-gram definition file(s) in ARPA
17       standard format to a compact Julius binary format. It will speed up the
18       initial loading time of N-gram much faster. It can read gzipped file
19       directly.
20
21       From rev.4.0, Julius can deal with forward N-gram, backward N-gram and
22       their combinations. So, mkbingram now generates binary N-gram file from
23       one of them, or combining them two to produce one binary N-gram.
24
25       When only a forward N-gram is specified, mkbingram generates binary
26       N-gram from only the forward N-gram. When using this binary N-gram at
27       Julius, it performs the 1st pass with the 2-gram probabilities in the
28       N-gram, and run the 2nd pass with the given N-gram fully, with
29       converting forward probabilities to backward probabilities by Bayes
30       rule.
31
32       When only a backward N-gram is specified, mkbingram generates an binary
33       N-gram file that contains only the backward N-gram. The 1st pass will
34       use forward 2-gram probabilities that can be computed from the backward
35       2-gram using Bayes rule, and the 2nd pass use the given backward N-gram
36       fully.
37
38       When both forward and backward N-grams are specified, the 2-gram part
39       in the forward N-gram and all backward N-gram will be combined into
40       single bingram file. The forward 2-gram will be applied for the 1st
41       pass and backward N-gram for the 2nd pass. Note that both N-gram should
42       be trained in the same corpus with same parameters (i.e. cut-off
43       thresholds), with same vocabulary.
44
45       The character code in binary N-gram can be converted from version 4.2.3
46       or later
47
48       The old binary N-gram produced by mkbingram of version 3.x and earlier
49       can be used in Julius-4, but you can convert the old version to the new
50       version by specifying it as input of current mkbingram by option "-d".
51
52       Please note that binary N-gram file converted by mkbingram of version
53       4.0 and later cannot be read by older Julius 3.x.
54

OPTIONS

56        -nlr  forward_ngram.arpa
57           Read in a forward (left-to-right) word N-gram file in ARPA standard
58           format.
59
60        -nrl  backward_ngram.arpa
61           Read in a backward (right-to-left) word N-gram file in ARPA
62           standard format.
63
64        -d  old_bingram_file
65           Read in a binary N-gram file.
66
67        -swap
68           Swap BOS word <s> and EOS word </s> in N-gram.
69
70        -c  from to
71           Convert character code in binary N-gram. ("from", "to" are string
72           that intend character code)
73
74       output_bingram_file
75           binary N-gram file name to output.
76

EXAMPLES

78       Convert a set of forward and backward N-gram in ARPA format into Julius
79       binary form:
80       Convert a single forward 4-gram in ARPA format into a binary file:
81       Convert old binary N-gram file to current format:
82

SEE ALSO

84        julius ( 1 ) ,
85        mkbinhmm ( 1 ) ,
86        mkbinhmmlist ( 1 )
87

89       Copyright (c) 1991-2013 Kawahara Lab., Kyoto University
90
91       Copyright (c) 1997-2000 Information-technology Promotion Agency, Japan
92
93       Copyright (c) 2000-2005 Shikano Lab., Nara Institute of Science and
94       Technology
95
96       Copyright (c) 2005-2013 Julius project team, Nagoya Institute of
97       Technology
98

LICENSE

100       The same as Julius.
101
102
103
104                                  12/19/2013                      MKBINGRAM(1)
Impressum