Upload app2.py

app2.py

@@ -0,0 +1,42 @@
+import os
+from speechbrain.pretrained.interfaces import foreign_class
+import gradio as gr
+
+import warnings
+warnings.filterwarnings("ignore")
+
+# Loading the speechbrain emotion detection model
+learner = foreign_class(
+    source="speechbrain/emotion-recognition-wav2vec2-IEMOCAP",
+    pymodule_file="custom_interface.py", 
+    classname="CustomEncoderWav2vec2Classifier"
+)
+
+# Building prediction function for gradio
+emotion_dict = {
+    'sad': 'Sad', 
+    'hap': 'Happy',
+    'ang': 'Anger',
+    'fea': 'Fear',
+    'sur': 'Surprised',
+    'neu': 'Neutral'
+}
+
+def predict_emotion(file_path):
+    # Since we get the file path from the dropdown, we don't need to access the `.name` property
+    out_prob, score, index, text_lab = learner.classify_file(file_path)
+    return emotion_dict[text_lab[0]]
+
+# Folder containing audio files
+folder = "prerecorded"
+
+# Assuming that the 'prerecorded' folder is in the current working directory
+# Change the working directory path if necessary
+audio_files = [os.path.join(folder, file) for file in os.listdir(folder) if file.endswith('.wav')]
+
+# Loading gradio interface with dropdown for audio selection
+inputs = gr.inputs.Dropdown(audio_files, label="Select Audio File")
+outputs = "text"
+title = "Machine Learning Emotion Detection"
+description = "Gradio demo for Emotion Detection. To use it, select an audio file from the dropdown and click 'Submit'. Read more at the links below."
+gr.Interface(predict_emotion, inputs, outputs, title=title, description=description).launch()