Update app.py

app.py

@@ -4,83 +4,85 @@ import soundfile as sf
 import os
 import base64
 import tempfile
+from fastapi import FastAPI, Request
+from fastapi.responses import JSONResponse
+import uvicorn
 
-# --- Model Loading ---
+# --- Load Model ---
 try:
     classifier = pipeline("audio-classification", model="superb/wav2vec2-base-superb-er")
 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
+    model_load_error = str(e)
+else:
+    model_load_error = None
 
-# --- Prediction Function ---
+# --- Gradio Prediction Function ---
 def predict_emotion(audio_file):
-    if classifier is None: 
-        return {"error": "The AI model could not be loaded."}
-    
-    if audio_file is None: 
+    if classifier is None:
+        return {"error": f"Model load failed: {model_load_error}"}
+    if audio_file is None:
         return {"error": "No audio input provided."}
-    
-    # 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)}"}
-    
+
     try:
+        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.wav"
+            sf.write(temp_audio_path, audio_array, sample_rate)
+            audio_path = temp_audio_path
+        else:
+            return {"error": f"Unsupported input type: {type(audio_file)}"}
+
         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)}"}
+    except Exception as e:
+        return {"error": f"Prediction error: {str(e)}"}
     finally:
-        if 'temp_audio_path' in locals() and os.path.exists(temp_audio_path): 
+        if 'temp_audio_path' in locals() and os.path.exists(temp_audio_path):
             os.remove(temp_audio_path)
 
-# --- API Function for Base64 Input ---
-def predict_emotion_api(data):
-    """
-    API function that accepts base64 encoded audio data
-    Expected input format: {"data": "base64_encoded_audio_string"}
-    """
+# --- FastAPI App for Base64 API ---
+app = FastAPI()
+
+@app.post("/api/predict/")
+async def predict_emotion_api(request: Request):
     if classifier is None:
-        return {"error": "The AI model could not be loaded."}
+        return JSONResponse(content={"error": f"Model load failed: {model_load_error}"}, status_code=500)
     
     try:
-        # Decode base64 audio data
-        audio_data = base64.b64decode(data)
-        
-        # Create temporary file
-        with tempfile.NamedTemporaryFile(delete=False, suffix='.wav') as temp_file:
+        body = await request.json()
+        base64_audio = body.get("data")
+        if not base64_audio:
+            return JSONResponse(content={"error": "Missing 'data' field with base64 audio."}, status_code=400)
+
+        audio_data = base64.b64decode(base64_audio)
+        with tempfile.NamedTemporaryFile(delete=False, suffix=".wav") as temp_file:
             temp_file.write(audio_data)
             temp_audio_path = temp_file.name
-        
-        # Predict emotion
+
         results = classifier(temp_audio_path, top_k=5)
-        
-        # Clean up temp file
         os.unlink(temp_audio_path)
-        
+
         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)}"}
+        return JSONResponse(content={"error": f"API prediction failed: {str(e)}"}, status_code=500)
 
-# --- Gradio Interface ---
-# Main interface for web UI
-iface = gr.Interface(
+# --- Gradio UI ---
+gradio_interface = 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.",
-    api_name="predict"  # This creates /api/predict/ endpoint
+    inputs=gr.Audio(sources=["microphone", "upload"], type="filepath", label="Upload Audio or Record"),
+    outputs=gr.Label(num_top_classes=5, label="Emotion Predictions"),
+    title="Audio Emotion Detector",
+    description="Upload or record your voice to detect emotions.",
+    allow_flagging="never"
 )
 
-# Launch the Gradio app
+# --- Mount Gradio inside FastAPI ---
+app = gr.mount_gradio_app(app, gradio_interface, path="/")
+
+# --- Launch for local/dev use only ---
 if __name__ == "__main__":
-    iface.queue().launch(server_name="0.0.0.0", share=True)
+    gradio_interface.queue()
+    uvicorn.run(app, host="0.0.0.0", port=7860)