[scripts,src] Check that symbol '#0' is not in the vocab of the ARPA LM file or the...

[processor-sdk/kaldi.git] / egs / wsj / s5 / utils / validate_dict_dir.pl

diff --git a/egs/wsj/s5/utils/validate_dict_dir.pl b/egs/wsj/s5/utils/validate_dict_dir.pl
index 5c089fcd077529cd97c847e63a5fb46b8ed16f68..a5c9ff8da21b519bc3aac69a875f6f88d6095304 100755 (executable)
--- a/egs/wsj/s5/utils/validate_dict_dir.pl
+++ b/egs/wsj/s5/utils/validate_dict_dir.pl
@@ -186,7 +186,7 @@ sub check_lexicon {
     if (!defined $word) {
       print "--> ERROR: empty lexicon line in $lex\n"; set_to_fail();
     }
     if (!defined $word) {
       print "--> ERROR: empty lexicon line in $lex\n"; set_to_fail();
     }

-    if ($word eq "<s>" || $word eq "</s>" || $word eq "<eps>") {
+    if ($word eq "<s>" || $word eq "</s>" || $word eq "<eps>" || $word eq "#0") {

       print "--> ERROR: lexicon.txt contains forbidden word $word\n";
       set_to_fail();
     }
       print "--> ERROR: lexicon.txt contains forbidden word $word\n";
       set_to_fail();
     }