Initial commit

app.py

@@ -0,0 +1,42 @@
+import gradio as gr
+import tensorflow as tf
+import librosa
+import numpy as np
+
+# Wczytanie modelu
+model = tf.keras.models.load_model("model.h5")
+
+# Lista klas (dostosuj do swojego przypadku)
+INSTRUMENTS = ['piano', 'guitar', 'violin', 'drums']  # <- zmień na własne klasy
+
+# Parametry przetwarzania
+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=predict_instrument,
+    inputs=gr.Audio(type="file", label="Wgraj plik WAV"),
+    outputs=gr.Label(num_top_classes=3),
+    title="Rozpoznawanie Instrumentów",
+    description="Model rozpoznaje instrumenty muzyczne na podstawie pliku dźwiękowego (.wav)."
+)
+
+if __name__ == "__main__":
+    demo.launch()

model.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:20cac95d5166f8f0b69e0d96d3dcc4e4f4228c59606c75b122f9b27e06c31c86
+size 281904

requirements.txt

@@ -0,0 +1,5 @@
+tensorflow
+librosa
+gradio
+numpy
+scikit-learn