Update app.py

app.py

@@ -147,7 +147,10 @@ def transcribe_video_with_speakers(video_path):
         # Transcribe
         result = model.transcribe(audio_path)
         logger.info("Audio transcription completed")
-    
+
+        # Get the detected language
+        detected_language = result["language"]
+        logger.debug(f"Detected language: {detected_language}")
         # Alignment
         model_a, metadata = whisperx.load_align_model(language_code=result["language"], device=device)
         result = whisperx.align(result["segments"], model_a, metadata, audio_path, device)
@@ -196,10 +199,6 @@ def transcribe_video_with_speakers(video_path):
         speaker_sample_paths[speaker] = sample_path
         logger.info(f"Created sample for {speaker}: {sample_path}")
 
-    # Get the detected language
-    detected_language = result["language"]
-    logger.debug(f"Detected language: {detected_language}")
-
     # Clean up
     video.close()
     audio_clip.close()