Update app.py

app.py

@@ -2,38 +2,119 @@ import gradio as gr
 from transformers import pipeline
 import soundfile as sf
 import os
+import logging
 
+# Set up logging to help debug issues
+logging.basicConfig(level=logging.INFO)
+logger = logging.getLogger(__name__)
+
+# Initialize the classifier with error handling
 try:
-    classifier = pipeline("audio-classification", model="superb/wav2vec2-base-superb-er")
+    # Using a more reliable emotion classification model
+    classifier = pipeline(
+        "audio-classification", 
+        model="ehcalabres/wav2vec2-lg-xlsr-en-speech-emotion-recognition",
+        return_all_scores=True
+    )
+    logger.info("Model loaded successfully")
 except Exception as e:
-    def error_fn(audio_file):
-        return {"error": f"Failed to load the model. Please check the logs. Error: {str(e)}"}
-    classifier = None
+    logger.error(f"Failed to load primary model: {e}")
+    try:
+        # Fallback to a different model
+        classifier = pipeline(
+            "audio-classification",
+            model="superb/wav2vec2-base-superb-er",
+            return_all_scores=True
+        )
+        logger.info("Fallback model loaded successfully")
+    except Exception as e2:
+        logger.error(f"Failed to load fallback model: {e2}")
+        classifier = None
 
 def predict_emotion(audio_file):
-    if classifier is None: return {"error": "The AI model could not be loaded."}
-    if audio_file is None: return {"error": "No audio input provided."}
-    if isinstance(audio_file, str): audio_path = audio_file
-    elif isinstance(audio_file, tuple):
-        sample_rate, audio_array = audio_file
-        temp_audio_path = "temp_audio_from_mic.wav"
-        sf.write(temp_audio_path, audio_array, sample_rate)
-        audio_path = temp_audio_path
-    else: return {"error": f"Invalid audio input format: {type(audio_file)}"}
+    """
+    Predict emotion from audio file
+    """
+    if classifier is None:
+        return {"error": "The AI model could not be loaded. Please check the logs."}
+    
+    if audio_file is None:
+        return {"error": "No audio input provided."}
+    
+    temp_audio_path = None
+    
     try:
-        results = classifier(audio_path, top_k=5)
-        return {item['label']: round(item['score'], 3) for item in results}
-    except Exception as e: return {"error": f"An error occurred during prediction: {str(e)}"}
+        # Handle different input types
+        if isinstance(audio_file, str):
+            audio_path = audio_file
+        elif isinstance(audio_file, tuple):
+            sample_rate, audio_array = audio_file
+            temp_audio_path = "temp_audio_from_mic.wav"
+            sf.write(temp_audio_path, audio_array, sample_rate)
+            audio_path = temp_audio_path
+        else:
+            return {"error": f"Invalid audio input format: {type(audio_file)}"}
+        
+        # Check if file exists
+        if not os.path.exists(audio_path):
+            return {"error": "Audio file not found"}
+        
+        # Perform emotion classification
+        logger.info(f"Processing audio file: {audio_path}")
+        results = classifier(audio_path)
+        
+        # Process results
+        if isinstance(results, list) and len(results) > 0:
+            # Sort by score and return top 5
+            sorted_results = sorted(results, key=lambda x: x['score'], reverse=True)[:5]
+            emotion_scores = {item['label']: round(item['score'], 3) for item in sorted_results}
+        else:
+            return {"error": "No valid results from the model"}
+        
+        logger.info(f"Prediction successful: {emotion_scores}")
+        return emotion_scores
+        
+    except Exception as e:
+        logger.error(f"Error during prediction: {str(e)}")
+        return {"error": f"An error occurred during prediction: {str(e)}"}
+    
     finally:
-        if 'temp_audio_path' in locals() and os.path.exists(temp_audio_path): os.remove(temp_audio_path)
+        # Clean up temporary file
+        if temp_audio_path and os.path.exists(temp_audio_path):
+            try:
+                os.remove(temp_audio_path)
+                logger.info("Temporary audio file cleaned up")
+            except Exception as e:
+                logger.warning(f"Failed to clean up temp file: {e}")
 
+# Create Gradio interface
 iface = gr.Interface(
     fn=predict_emotion,
-    inputs=gr.Audio(sources=["microphone", "upload"], type="filepath", label="Upload Audio or Record with Microphone"),
-    outputs=gr.Label(num_top_classes=5, label="Emotion Probabilities"),
-    title="AI Audio Emotion Detector",
-    description="Upload an audio file or record your voice to detect emotions.",
+    inputs=gr.Audio(
+        sources=["microphone", "upload"], 
+        type="filepath", 
+        label="Upload Audio or Record with Microphone"
+    ),
+    outputs=gr.Label(
+        num_top_classes=5, 
+        label="Emotion Probabilities"
+    ),
+    title="🎵 AI Audio Emotion Detector",
+    description="Upload an audio file or record your voice to detect emotions. Supported formats: WAV, MP3, M4A, FLAC.",
+    article="This tool uses advanced AI models to analyze emotional content in speech and audio.",
+    examples=None,  # You can add example audio files here if you have them
+    allow_flagging="never"
 )
 
 if __name__ == "__main__":
-    iface.queue().launch()
+    try:
+        # Launch with queue for better handling of concurrent requests
+        iface.queue(max_size=10).launch(
+            server_name="0.0.0.0",  # Allow external access
+            server_port=7860,       # Default Gradio port
+            share=True,             # Create a public link
+            debug=True              # Enable debug mode
+        )
+    except Exception as e:
+        logger.error(f"Failed to launch Gradio app: {e}")
+        print(f"Error launching app: {e}")