add bird sound classification app and update .gitattributes

.gitattributes

@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+*.mp3 filter=lfs diff=lfs merge=lfs -text
+*.wav filter=lfs diff=lfs merge=lfs -text

.gitignore

@@ -0,0 +1 @@
+.idea

app.py

@@ -0,0 +1,47 @@
+import gradio as gr
+import torch
+import librosa
+from transformers import Wav2Vec2ForSequenceClassification, Wav2Vec2FeatureExtractor
+
+# Load model from Hugging Face Hub
+model_name = "greenarcade/wav2vec2-vd-bird-sound-classification"
+model = Wav2Vec2ForSequenceClassification.from_pretrained(model_name)
+feature_extractor = Wav2Vec2FeatureExtractor.from_pretrained(model_name)
+
+
+def predict(audio_file):
+    # Handle MP3/WAV files
+    audio, sr = librosa.load(audio_file, sr=16000)
+
+    # Process audio
+    inputs = feature_extractor(
+        audio,
+        sampling_rate=16000,
+        return_tensors="pt",
+        padding=True,
+        truncation=True,
+        max_length=16000 * 5,
+    )
+
+    # Predict
+    with torch.no_grad():
+        logits = model(**inputs).logits
+    probs = torch.softmax(logits, dim=-1).squeeze().tolist()
+
+    # Format results - return actual float values instead of formatted strings
+    predictions = {model.config.id2label[i]: prob for i, prob in enumerate(probs)}
+    sorted_preds = sorted(predictions.items(), key=lambda x: x[1], reverse=True)[:3]
+    return {k: v for k, v in sorted_preds}
+
+
+# Gradio Interface
+demo = gr.Interface(
+    fn=predict,
+    inputs=gr.Audio(sources=["upload"], type="filepath"),
+    outputs=gr.Label(num_top_classes=3),
+    title="🦜 Bird Sound Classifier (Indian birds)",
+    description="Upload a 5-second audio clip to identify bird species",
+    examples=[["greyheron-sample.wav"], ["blue-tail-sample.mp3"]]
+)
+
+demo.launch()