gp_norm_dict_CZ.pl
3.64 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
#!/usr/bin/env perl
use warnings; #sed replacement for -w perl parameter
# Copyright 2012 Arnab Ghoshal; Milos Janda
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# THIS CODE IS PROVIDED *AS IS* BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
# KIND, EITHER EXPRESS OR IMPLIED, INCLUDING WITHOUT LIMITATION ANY IMPLIED
# WARRANTIES OR CONDITIONS OF TITLE, FITNESS FOR A PARTICULAR PURPOSE,
# MERCHANTABLITY OR NON-INFRINGEMENT.
# See the Apache 2 License for the specific language governing permissions and
# limitations under the License.
# This script normalizes the GlobalPhone Czech dictionary. It (optionally)
# tags the phones with language ('CZ') marker. It also converts the words to
# UTF8 and lowercases everything, either of which can be diabled with command
# line switches.
# No special treatment for acronyms since there is no easy way of detecting
# acronyms in the dictionary.
my $usage = "Usage: gp_norm_dict_CZ.pl [-l|-m map|-u] -i dictionary > formatted\
Normalizes pronunciation dictionary for GlobalPhone Czech.\
There will probably be duplicates; so pipe the output through sort -u \
Options:\
-l\t\tAdd language tag to the phones
-m FILE\tMapping to a different phoneset
-u\t\tConvert words to uppercase (by default make everything lowercase)\n";
use strict;
use Getopt::Long;
use Unicode::Normalize;
use open ':encoding(iso-8859-2)';
binmode(STDOUT, ":encoding(utf8)");
die "$usage" unless(@ARGV >= 1);
my ($in_dict, $lang_tag, $map_file, $uppercase);
GetOptions ("l" => \$lang_tag, # tag phones with language ID.
"m=s" => \$map_file, # map to a different phoneset
"u" => \$uppercase, # convert words to uppercase
"i=s" => \$in_dict); # Input lexicon
my %phone_map = ();
if (defined($map_file)) {
warn "Language tag added (-l) while mapping to different phoneset (-m)"
if (defined($lang_tag));
open(M, "<$map_file") or die "Cannot open phone mapping file '$map_file': $!";
while (<M>) {
next if /^\#/; # Skip comments
s/\r//g; # Since files may have CRLF line-breaks!
chomp;
next if /^$/; # skip empty lines
# The mapping is assumed to be: 'from-phone' 'to-phone'
die "Bad line: $_" unless m/^(\S+)\s+(\S+).*$/;
die "Multiple mappings for phone $1: '$2' and '$phone_map{$1}'"
if (defined($phone_map{$1}));
$phone_map{$1} = $2;
}
}
open(L, "<$in_dict") or die "Cannot open dictionary file '$in_dict': $!";
while (<L>) {
s/\r//g; # Since files may have CRLF line-breaks!
chomp;
$_ =~ m:^\{?(\S*?)\}?\s+\{?(.+?)\}?$: or die "Bad line: $_";
my $word = $1;
my $pron = $2;
next if ($pron =~ /SIL/); # Silence will be added later to the lexicon
# First, normalize the pronunciation:
$pron =~ s/\{//g;
$pron =~ s/^\s*//; $pron =~ s/\s*$//; # remove leading or trailing spaces
$pron =~ s/ WB\}//g;
$pron =~ s/\s+/ /g; # Normalize spaces
$pron =~ s/M_//g; # Get rid of the M_ marker before the phones
if (defined($map_file)) {
my (@phones) = split(' ', $pron);
for my $i (1..$#phones) {
if (defined($phone_map{$phones[$i]})) {
$phones[$i] = $phone_map{$phones[$i]};
} else {
warn "No mapping found for $phones[$i]: keeping original.";
}
}
$pron = join(' ', @phones);
}
$pron =~ s/(\S+)/$1_CZ/g if(defined($lang_tag));
# Next, normalize the word:
$word =~ s/\(.*\)//g; # Pron variants should have same orthography
if (defined($uppercase)) {
$word = uc($word);
} else {
$word = lc($word);
}
print "$word\t$pron\n";
}
close(L);