commit1

app.py

@@ -14,9 +14,24 @@ SR = 22050
 DURATION = 5
 N_MELS = 128
 
+def preprocess_audio(file):
+    y, _ = librosa.load(file.name, sr=SR, duration=DURATION)
+    if len(y) < SR * DURATION:
+        y = np.pad(y, (0, SR * DURATION - len(y)))
+    mel = librosa.feature.melspectrogram(y=y, sr=SR, n_mels=N_MELS)
+    mel_db = librosa.power_to_db(mel, ref=np.max)
+    mel_db = mel_db[..., np.newaxis]  # Dodaj kanał
+    mel_db = np.expand_dims(mel_db, axis=0)  # Dodaj batch
+    return mel_db
+
+def predict_instrument(audio_file):
+    mel_input = preprocess_audio(audio_file)
+    preds = model.predict(mel_input)[0]
+    result = {cls: float(score) for cls, score in zip(INSTRUMENTS, preds)}
+    return result
 
 demo = gr.Interface(
-    fn=print("F"),
+    fn=predict_instrument,
     inputs=gr.Audio(type="filepath", label="Wgraj plik WAV"),
     outputs=gr.Label(num_top_classes=3),
     title="Rozpoznawanie Instrumentów",