| MKBINGRAM(1) | MKBINGRAM(1) |
mkbingram
- make binary N-gram from ARPA N-gram file
mkbingram [-nlr forward_ngram.arpa] [-nrl backward_ngram.arpa] [-d old_bingram_file] {output_bingram_file}
mkbingram is a tool to convert N-gram definition file(s) in ARPA standard format to a compact Julius binary format. It will speed up the initial loading time of N-gram much faster. It can read gzipped file directly.
From rev.4.0, Julius can deal with forward N-gram, backward N-gram and their combinations. So, mkbingram now generates binary N-gram file from one of them, or combining them two to produce one binary N-gram.
When only a forward N-gram is specified, mkbingram generates binary N-gram from only the forward N-gram. When using this binary N-gram at Julius, it performs the 1st pass with the 2-gram probabilities in the N-gram, and run the 2nd pass with the given N-gram fully, with converting forward probabilities to backward probabilities by Bayes rule.
When only a backward N-gram is specified, mkbingram generates an binary N-gram file that contains only the backward N-gram. The 1st pass will use forward 2-gram probabilities that can be computed from the backward 2-gram using Bayes rule, and the 2nd pass use the given backward N-gram fully.
When both forward and backward N-grams are specified, the 2-gram part in the forward N-gram and all backward N-gram will be combined into single bingram file. The forward 2-gram will be applied for the 1st pass and backward N-gram for the 2nd pass. Note that both N-gram should be trained in the same corpus with same parameters (i.e. cut-off thresholds), with same vocabulary.
The old binary N-gram produced by mkbingram of version 3.x and earlier can be used in Julius-4, but you can convert the old version to the new version by specifying it as input of current mkbingram by option "-d".
Please note that binary N-gram file converted by mkbingram of version 4.0 and later cannot be read by older Julius 3.x.
-nlr forward_ngram.arpa
-nrl backward_ngram.arpa
-d old_bingram_file
-swap
output_bingram_file
Convert a set of forward and backward N-gram in ARPA format into Julius binary form:
Convert a single forward 4-gram in ARPA format into a binary file:
Convert old binary N-gram file to current format:
julius ( 1 ) , mkbinhmm ( 1 ) , mkbinhmmlist ( 1 )
Copyright (c) 1997-2000 Information-technology Promotion Agency, Japan
Copyright (c) 1991-2008 Kawahara Lab., Kyoto University
Copyright (c) 2000-2005 Shikano Lab., Nara Institute of Science and Technology
Copyright (c) 2005-2008 Julius project team, Nagoya Institute of Technology
The same as Julius.
| 02/11/2009 |