DebasishDhal99 commited on
Commit
718e9ec
1 Parent(s): 36e9b92

Debugging Turkish end char pronunciation

Browse files
Files changed (1) hide show
  1. turkish.py +9 -3
turkish.py CHANGED
@@ -1,6 +1,7 @@
1
  # import nltk
2
  # nltk.download('punkt')
3
  # from nltk.tokenize import word_tokenize
 
4
 
5
  special_combs = {"c" : "ј", "C" : "Ј"} #These are Serbian J characters, they will be later converted to Latin J.
6
 
@@ -44,15 +45,20 @@ def check_punc(word): #The pronunciation of some Turkish chars change if they're
44
  punc_list = list(string.punctuation)
45
 
46
  ending_punc = False
47
- #Modify it such that even it there are >1 puncutaiton marks at the word ending, it'll detect that
 
 
48
  for punc in punc_list:
49
  if word.endswith(punc):
50
  ending_punc = True
51
-
52
- how_many_punc_end = 0
53
  for i in range(1,len(word)):
54
  if word[-i] in punc_list:
55
  how_many_punc_end += 1
 
 
 
56
  return ending_punc, how_many_punc_end
57
 
58
 
 
1
  # import nltk
2
  # nltk.download('punkt')
3
  # from nltk.tokenize import word_tokenize
4
+ import string
5
 
6
  special_combs = {"c" : "ј", "C" : "Ј"} #These are Serbian J characters, they will be later converted to Latin J.
7
 
 
45
  punc_list = list(string.punctuation)
46
 
47
  ending_punc = False
48
+ how_many_punc_end = 0
49
+ #Modify it such that even it there are > 1 puncutaiton marks at the word ending, it'll detect that
50
+
51
  for punc in punc_list:
52
  if word.endswith(punc):
53
  ending_punc = True
54
+ break
55
+
56
  for i in range(1,len(word)):
57
  if word[-i] in punc_list:
58
  how_many_punc_end += 1
59
+
60
+ if word[-i] not in punc_list:
61
+ break
62
  return ending_punc, how_many_punc_end
63
 
64