[tn] delete prefix space (#262)

wenet-e2e · Jul 4, 2024 · 2b04bc3 · 2b04bc3
1 parent 91e51ca
commit 2b04bc3
Show file tree

Hide file tree

Showing 2 changed files with 4 additions and 2 deletions.
diff --git a/tn/english/normalizer.py b/tn/english/normalizer.py
@@ -73,8 +73,9 @@ def build_tagger(self):
              | rang
              | punct
              ).optimize() + (punct.plus | self.DELETE_SPACE)
-        # delete the last space
-        self.tagger = tagger.star @ self.build_rule(delete(' '), r='[EOS]')
+        # delete the first and last space
+        self.tagger = (delete(' ').star + tagger.star) @ self.build_rule(
+            delete(' '), r='[EOS]')
 
     def build_verbalizer(self):
         cardinal = Cardinal().verbalizer

diff --git a/tn/english/test/data/normalizer.txt b/tn/english/test/data/normalizer.txt
@@ -5,3 +5,4 @@ The museum is open Mon.-Sun. children of 3-4 years 123 The plan will help you lo
 Try searching for 'Toyota' or 'Investment' => Try searching for 'Toyota' or 'Investment'
 "" => ""
 The HTML tag <p> defines a paragraph. => The HTML tag <p> defines a paragraph.
+ hello world => hello world