Create app.py

app.py

@@ -0,0 +1,25 @@
+import gradio as gr
+import torch
+import librosa
+from transformers import Wav2Vec2ForSequenceClassification, Wav2Vec2FeatureExtractor
+
+# Load the model and feature extractor
+model_name = "r-f/wav2vec-english-speech-emotion-recognition"
+model = Wav2Vec2ForSequenceClassification.from_pretrained(model_name)
+feature_extractor = Wav2Vec2FeatureExtractor.from_pretrained(model_name)
+
+# Define the emotion labels
+labels = ['angry', 'disgust', 'fear', 'happy', 'neutral', 'sad', 'surprise']
+
+def predict_emotion(audio):
+    # Load and preprocess the audio
+    audio, rate = librosa.load(audio, sr=16000)
+    inputs = feature_extractor(audio, sampling_rate=rate, return_tensors="pt", padding=True)
+    with torch.no_grad():
+        logits = model(**inputs).logits
+    predicted_class_id = torch.argmax(logits).item()
+    return labels[predicted_class_id]
+
+# Create the Gradio interface
+interface = gr.Interface(fn=predict_emotion, inputs=gr.Audio(type="filepath"), outputs="text")
+interface.launch()