Update utils.py

utils.py

@@ -490,7 +490,7 @@ def _preprocess_text_for_tts(text: str, speaker: str) -> str:
     """
     1) "SaaS" => "sass"
     2) Insert periods for uppercase abbreviations -> remove for TTS
-    3) Convert decimals like "3.14" -> "three point one four"
+    3) Preserve numbers for natural TTS pronunciation
     4) Expand leftover all-caps
     5) Emotive placeholders for 'ha', 'haha', 'sigh', 'groan', etc.
     6) If speaker != Jane, insert filler words
@@ -512,17 +512,18 @@ def _preprocess_text_for_tts(text: str, speaker: str) -> str:
         return chunk.replace(".", " ").strip()
     text = re.sub(r"[A-Z0-9]\.[A-Z0-9](?:\.[A-Z0-9])*\.", remove_periods_for_tts, text)
 
-    # 3) Hyphens -> spaces
-    text = re.sub(r"-", " ", text)
+    # 3) Preserve numbers by removing any digit-specific processing
+    # Let TTS handle natural number pronunciation
 
-    # Removed numeric conversions to let TTS handle numbers naturally.
+    # 4) Hyphens -> spaces (but preserve hyphenated numbers)
+    text = re.sub(r"(?<!\d)-(?!\d)", " ", text)
 
-    # 6) Emotive placeholders
+    # 5) Emotive placeholders
     text = re.sub(r"\b(ha(ha)?|heh|lol)\b", "(* laughs *)", text, flags=re.IGNORECASE)
     text = re.sub(r"\bsigh\b", "(* sighs *)", text, flags=re.IGNORECASE)
     text = re.sub(r"\b(groan|moan)\b", "(* groans *)", text, flags=re.IGNORECASE)
 
-    # 7) Insert filler words if speaker != "Jane"
+    # 6) Insert filler words if speaker != "Jane"
     if speaker != "Jane":
         def insert_thinking_pause(m):
             word = m.group(1)
@@ -537,10 +538,10 @@ def _preprocess_text_for_tts(text: str, speaker: str) -> str:
         conj_pattern = r"\b(and|but|so|because|however)\b"
         text = re.sub(conj_pattern, lambda m: f"{m.group()}...", text, flags=re.IGNORECASE)
 
-    # 8) Remove random fillers
+    # 7) Remove random fillers
     text = re.sub(r"\b(uh|um|ah)\b", "", text, flags=re.IGNORECASE)
 
-    # 9) Capitalize sentence starts
+    # 8) Capitalize sentence starts
     def capitalize_match(m):
         return m.group().upper()
     text = re.sub(r'(^\s*\w)|([.!?]\s*\w)', capitalize_match, text)