Update app.py

app.py

@@ -48,59 +48,70 @@ async def process_transcript_line(line, voice, rate, pitch):
         if not text_with_commands.strip():
             return start_time_ms, None
 
-        current_voice = voice
+        current_voice_full = voice  # Store the full voice string from the dropdown
+        current_voice_short = current_voice_full.split(" - ")[0] if current_voice_full else ""
         current_rate = rate
         current_pitch = pitch
         processed_text = text_with_commands
 
-        voice1 = "en-AU-WilliamNeural - en-AU (Male)"
-        voice1F ="en-GB-SoniaNeural - en-GB (Female)"
-        voice2 = "en-GB-RyanNeural - en-GB (Male)"
-        voice2F = "en-US-JennyNeural - en-US (Female)"
-        voice3 ="en-US-BrianMultilingualNeural - en-US (Male)"  #good for reading
-        voice3F = "en-HK-YanNeural - en-HK (Female)"
-        voice4 = "en-GB-ThomasNeural - en-GB (Male)"
-        voice4F ="en-US-EmmaNeural - en-US (Female)"
-        voice5 = "en-GB-RyanNeural - en-GB (Male)" #Old Man
-        voice6 = "en-GB-MaisieNeural - en-GB (Female)"  #Child
+        voice1_full = "en-AU-WilliamNeural - en-AU (Male)"
+        voice1_short = voice1_full.split(" - ")[0]
+        voice1F_full ="en-GB-SoniaNeural - en-GB (Female)"
+        voice1F_short = voice1F_full.split(" - ")[0]
+        voice2_full = "en-GB-RyanNeural - en-GB (Male)"
+        voice2_short = voice2_full.split(" - ")[0]
+        voice2F_full = "en-US-JennyNeural - en-US (Female)"
+        voice2F_short = voice2F_full.split(" - ")[0]
+        voice3_full ="en-US-BrianMultilingualNeural - en-US (Male)"  #good for reading
+        voice3_short = voice3_full.split(" - ")[0]
+        voice3F_full = "en-HK-YanNeural - en-HK (Female)"
+        voice3F_short = voice3F_full.split(" - ")[0]
+        voice4_full = "en-GB-ThomasNeural - en-GB (Male)"
+        voice4_short = voice4_full.split(" - ")[0]
+        voice4F_full ="en-US-EmmaNeural - en-US (Female)"
+        voice4F_short = voice4F_full.split(" - ")[0]
+        voice5_full = "en-GB-RyanNeural - en-GB (Male)" #Old Man
+        voice5_short = voice5_full.split(" - ")[0]
+        voice6_full = "en-GB-MaisieNeural - en-GB (Female)"  #Child
+        voice6_short = voice6_full.split(" - ")[0]
 
         if text_with_commands.startswith("1F"):
-            current_voice = voice1F.split(" - ")[0]
+            current_voice_short = voice1F_short
             current_pitch = 25
             processed_text = text_with_commands[2:].strip()
         elif text_with_commands.startswith("2F"):
-            current_voice = voice2F.split(" - ")[0]
+            current_voice_short = voice2F_short
             processed_text = text_with_commands[2:].strip()
         elif text_with_commands.startswith("3F"):
-            current_voice = voice3F.split(" - ")[0]
+            current_voice_short = voice3F_short
             processed_text = text_with_commands[2:].strip()
         elif text_with_commands.startswith("4F"):
-            current_voice = voice4F.split(" - ")[0]
+            current_voice_short = voice4F_short
             processed_text = text_with_commands[2:].strip()
         elif text_with_commands.startswith("1M"):
-            current_voice = voice1.split(" - ")[0]
+            current_voice_short = voice1_short
             processed_text = text_with_commands[2:].strip()
         elif text_with_commands.startswith("2M"):
-            current_voice = voice2.split(" - ")[0]
+            current_voice_short = voice2_short
             processed_text = text_with_commands[2:].strip()
         elif text_with_commands.startswith("3M"):
-            current_voice = voice3.split(" - ")[0]
+            current_voice_short = voice3_short
             processed_text = text_with_commands[2:].strip()
         elif text_with_commands.startswith("4M"):
-            current_voice = voice4.split(" - ")[0]
+            current_voice_short = voice4_short
             processed_text = text_with_commands[2:].strip()
         elif text_with_commands.startswith("1O"):  # Old man voice
-            current_voice = voice5.split(" - ")[0]
+            current_voice_short = voice5_short
             current_pitch = -20
             current_rate = -10
             processed_text = text_with_commands[2:].strip()
         elif text_with_commands.startswith("1C"):  #Child voice
-            current_voice = voice6.split(" - ")[0]
+            current_voice_short = voice6_short
             processed_text = text_with_commands[2:].strip()
 
         rate_str = f"{current_rate:+d}%"
         pitch_str = f"{current_pitch:+d}Hz"
-        communicate = edge_tts.Communicate(processed_text, current_voice, rate=rate_str, pitch=pitch_str)
+        communicate = edge_tts.Communicate(processed_text, current_voice_short, rate=rate_str, pitch=pitch_str)
         with tempfile.NamedTemporaryFile(delete=False, suffix=".mp3") as tmp_file:
             audio_path = tmp_file.name
             await communicate.save(audio_path)