Update app.py

app.py

@@ -40,20 +40,15 @@ import wave
 
 logger = logging.getLogger(__name__)
 
+# Configure logging
+logging.basicConfig(level=logging.DEBUG, format="%(asctime)s - %(levelname)s - %(message)s")
+logger = logging.getLogger(__name__)
+logger.info(f"MoviePy Version: {moviepy.__version__}")
+
 # Accept license terms for Coqui XTTS
 os.environ["COQUI_TOS_AGREED"] = "1"
 # torch.serialization.add_safe_globals([XttsConfig])
 
-# Load XTTS model
-try:
-    print("🔄 Loading XTTS model...")
-    tts = TTS(model_name="tts_models/multilingual/multi-dataset/xtts_v2")
-    print("✅ XTTS model loaded successfully.")
-except Exception as e:
-    print("❌ Error loading XTTS model:")
-    traceback.print_exc()
-    raise e
-
 logger.info(gr.__version__)
 
 client = OpenAI(
@@ -61,7 +56,6 @@ client = OpenAI(
 )
 hf_api_key = os.environ.get("hf_token")
 
-
 def silence(duration, fps=44100):
     """
     Returns a silent AudioClip of the specified duration.
@@ -132,11 +126,6 @@ def handle_feedback(feedback):
             conn.commit()
         return "Thank you for your feedback!", None
 
-# Configure logging
-logging.basicConfig(level=logging.DEBUG, format="%(asctime)s - %(levelname)s - %(message)s")
-logger = logging.getLogger(__name__)
-logger.info(f"MoviePy Version: {moviepy.__version__}")
-
 # def segment_background_audio(audio_path, output_path="background_segments.wav"):
 #     # Step 2: Initialize pyannote voice activity detection pipeline (you need Hugging Face token)
 #     pipeline = Pipeline.from_pretrained(
@@ -490,9 +479,6 @@ def add_transcript_voiceover(video_path, translated_json, output_path, add_voice
 
     return error_messages
 
-# Initialize TTS model only once (outside the function)
-tts = TTS(model_name="tts_models/multilingual/multi-dataset/xtts_v2")
-
 def generate_voiceover_clone(translated_json, desired_duration, target_language, speaker_wav_path, output_audio_path):
     try:
         full_text = " ".join(entry["translated"] for entry in translated_json if "translated" in entry and entry["translated"].strip())
@@ -677,6 +663,16 @@ def build_interface():
 
     return demo
 
+    # Load XTTS model
+try:
+    print("🔄 Loading XTTS model...")
+    tts = TTS(model_name="tts_models/multilingual/multi-dataset/xtts_v2")
+    print("✅ XTTS model loaded successfully.")
+except Exception as e:
+    print("❌ Error loading XTTS model:")
+    traceback.print_exc()
+    raise e
+    
 # Launch the Gradio interface
 demo = build_interface()
-demo.launch()
+demo.launch()