U

app.py

@@ -1,18 +1,26 @@
 import gradio as gr
 from transformers import pipeline
 
-# Load model
-classifier = pipeline("audio-classification", model="/spaces/Hnin/Audio_Classification_On_Key_spotting")
+# Load model - use the correct Hugging Face model ID
+# Remove the '/spaces/' prefix and use just the username/model-name format
+classifier = pipeline("audio-classification", model="Hnin/Audio_Classification_On_Key_spotting")
 
 def predict(audio):
-    preds = classifier(audio)
-    return {p["label"]: p["score"] for p in preds}
+    if audio is None:
+        return {"Error": "No audio provided"}
+    
+    try:
+        preds = classifier(audio)
+        return {p["label"]: p["score"] for p in preds}
+    except Exception as e:
+        return {"Error": f"Prediction failed: {str(e)}"}
 
 # Gradio UI
 gr.Interface(
     fn=predict,
-    inputs=gr.Audio(source="microphone", type="filepath"),
+    inputs=gr.Audio(sources=["microphone"], type="filepath"),  # Updated parameter name
     outputs=gr.Label(num_top_classes=3),
     title="🔊 Keyword Spotting",
-    examples=["example1.wav", "example2.wav"]
+    description="Upload an audio file or record from microphone for keyword spotting classification",
+    examples=["example1.wav", "example2.wav"]  # Make sure these files exist
 ).launch()