Add removes weird characters from text string such as underscore, asterix, etc

app.py

@@ -39,6 +39,23 @@ async def get_voices():
     voices = await edge_tts.list_voices()
     return {f"{v['ShortName']} - {v['Locale']} ({v['Gender']})": v['ShortName'] for v in voices}
 
+def clean_text(text: str) -> str:
+    """
+    Cleans a text string by:
+    - Removing unwanted characters (except letters, digits, spaces, commas, periods)
+    - Fixing broken words caused by dots and spaces
+    - Normalizing spaces
+    """
+    # Remove unwanted characters
+    text = re.sub(r"[^a-zA-Z0-9\s,\.]", '', text)
+    # Fix broken words (remove dots and spaces between letters/digits)
+    text = re.sub(r'(?<=\w)[\.\s]+(?=\w)', '', text)
+    # Normalize multiple spaces to one
+    text = re.sub(r"\s+", ' ', text)
+    # Trim leading and trailing spaces
+    text = text.strip()
+    return text    
+
 # Text-to-speech function for a single paragraph with SS handling
 async def paragraph_to_speech(text, voice, rate, pitch):
     voice1 = "en-AU-WilliamNeural - en-AU (Male)"    
@@ -138,7 +155,8 @@ async def paragraph_to_speech(text, voice, rate, pitch):
             #    processed_text = part[4:]
             #    pitch = int(part[2:4])
             pitch_str = f"{current_pitch:+d}Hz"
-            communicate = edge_tts.Communicate(processed_text, current_voice, rate=rate_str, pitch=pitch_str)
+            texttosend = clean_text (processed_text)
+            communicate = edge_tts.Communicate(texttosend, current_voice, rate=rate_str, pitch=pitch_str)
             with tempfile.NamedTemporaryFile(delete=False, suffix=".mp3") as tmp_file:
                 tmp_path = tmp_file.name
                 await communicate.save(tmp_path)