Create app.py

app.py

@@ -0,0 +1,27 @@
+import gradio as gr
+import torch
+from transformers import Wav2Vec2ForSequenceClassification, Wav2Vec2Processor
+
+# Load the model and processor from Hugging Face
+model = Wav2Vec2ForSequenceClassification.from_pretrained("HareemFatima/distilhubert-finetuned-stutterdetection")
+processor = Wav2Vec2Processor.from_pretrained("HareemFatima/distilhubert-finetuned-stutterdetection")
+
+# Define a function for stutter detection
+def detect_stutter(audio):
+    # Preprocess the audio
+    inputs = processor(audio, sampling_rate=16000, return_tensors="pt", padding=True)
+
+    # Get model predictions
+    with torch.no_grad():
+        logits = model(**inputs).logits
+    predicted_class = logits.argmax(-1).item()
+    
+    # Map prediction to stutter type
+    stutter_types = {0: "Non Stutter", 1: "Beginner Stutter", 2: "Middle Stutter", 3: "End Stutter"}
+    return stutter_types.get(predicted_class, "Unknown Stutter")
+
+# Create Gradio interface
+iface = gr.Interface(fn=detect_stutter, inputs=gr.Audio(source="microphone", type="numpy"), outputs="text")
+
+# Launch the interface
+iface.launch()