Create app.py

app.py

@@ -0,0 +1,51 @@
+import gradio as gr
+import numpy as np
+import librosa
+import onnxruntime as rt
+from transformers import Pipeline
+
+class MujawwadPipeline:
+    def __init__(self, model_path='model/mujawwad_classifier.onnx'):
+        self.session = rt.InferenceSession(model_path)
+        self.labels = ['bayati','hijaz','jiharkah','nihawand','rast','shoba','sikah']
+        self.descriptions = {
+            'bayati': 'Bayati (بياتي)',
+            'hijaz': 'Hijaz (حجاز)',
+            'jiharkah': 'Jiharkah (جهاركاه)',
+            'nihawand': 'Nihawand (نهاوند)',
+            'rast': 'Rast (راست)',
+            'shoba': 'Shoba (صبا)',
+            'sikah': 'Sikah (سيكاه)'
+        }
+
+    def predict(self, audio_path):
+        try:
+            input_data = self.preprocess(audio_path)
+            input_name = self.session.get_inputs()[0].name
+            output_name = self.session.get_outputs()[0].name
+            predictions = self.session.run([output_name], {input_name: input_data.astype(np.float32)})[0]
+            
+            # Format results with descriptions
+            results = {self.descriptions[self.labels[i]]: float(predictions[0][i]) 
+                      for i in range(len(self.labels))}
+            return results
+        except Exception as e:
+            return {"Error": str(e)}
+
+def classify_audio(audio_path):
+    classifier = MujawwadPipeline()
+    result = classifier.predict(audio_path)
+    return result
+
+# Create Gradio interface
+iface = gr.Interface(
+    fn=classify_audio,
+    inputs=gr.Audio(type="filepath", label="Upload Quranic Recitation"),
+    outputs=gr.Label(num_top_classes=7),
+    title="Maqamat Classification Model",
+    description="Classify Quranic recitation maqamat into 7 traditional Arabic melodic modes",
+    examples=[["examples/bayati_example.mp3"], ["examples/hijaz_example.mp3"]]
+)
+
+if __name__ == "__main__":
+    iface.launch()