HusseinBashir commited on
Commit
0244582
·
verified ·
1 Parent(s): 38b2d32

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +9 -3
app.py CHANGED
@@ -85,6 +85,14 @@ def normalize_text(text):
85
  # ➤ Beddel ereyga zamzam dhammaan noocyadiisa (bilow, dhex, dhammaad)
86
  text = re.sub(r'(?i)(?<!\w)zamzam(?!\w)', 'samsam', text)
87
 
 
 
 
 
 
 
 
 
88
  # ➤ Ka saar tirooyin leh koma iyo tobanle
89
  text = re.sub(r'(\d{1,3})(,\d{3})+', lambda m: m.group(0).replace(",", ""), text)
90
  text = re.sub(r'\.\d+', '', text)
@@ -108,14 +116,12 @@ def normalize_text(text):
108
  text = text.replace("KH", "qa").replace("Z", "S")
109
  text = text.replace("SH", "SHa'a").replace("DH", "Dha'a")
110
 
111
- # ➤ Haddii jumladda ku dhammaato zamzam ama samsam, ku dar " ah"
112
  if re.search(r'(?i)(zamzam|samsam)[\s\.\,\!\?]*$', text.strip()):
113
  text += " m"
114
 
115
  return text
116
 
117
-
118
-
119
  def tts(text):
120
  paragraphs = text.strip().split("\n")
121
  audio_list = []
 
85
  # ➤ Beddel ereyga zamzam dhammaan noocyadiisa (bilow, dhex, dhammaad)
86
  text = re.sub(r'(?i)(?<!\w)zamzam(?!\w)', 'samsam', text)
87
 
88
+ # ➤ Bedel shortcuts - eray kasta oo qoraalka ku jira beddel
89
+ # Ka dhig case-insensitive beddelka
90
+ def replace_shortcuts(match):
91
+ word = match.group(0).lower()
92
+ return shortcut_map.get(word, word)
93
+ pattern = re.compile(r'\b(' + '|'.join(re.escape(k) for k in shortcut_map.keys()) + r')\b', re.IGNORECASE)
94
+ text = pattern.sub(replace_shortcuts, text)
95
+
96
  # ➤ Ka saar tirooyin leh koma iyo tobanle
97
  text = re.sub(r'(\d{1,3})(,\d{3})+', lambda m: m.group(0).replace(",", ""), text)
98
  text = re.sub(r'\.\d+', '', text)
 
116
  text = text.replace("KH", "qa").replace("Z", "S")
117
  text = text.replace("SH", "SHa'a").replace("DH", "Dha'a")
118
 
119
+ # ➤ Haddii jumladda ku dhammaato zamzam ama samsam, ku dar " m" (silent m)
120
  if re.search(r'(?i)(zamzam|samsam)[\s\.\,\!\?]*$', text.strip()):
121
  text += " m"
122
 
123
  return text
124
 
 
 
125
  def tts(text):
126
  paragraphs = text.strip().split("\n")
127
  audio_list = []