4

app4.py

@@ -30,34 +30,27 @@ emotion_dict = {
 }
 
 def predict_emotion(selected_audio):
-    if selected_audio is None:  # Check if an audio file is selected
-        return "Please select an audio file.", None
     file_path = os.path.join("rec", selected_audio)
     out_prob, score, index, text_lab = learner.classify_file(file_path)
     emotion = emotion_dict[text_lab[0]]
     return emotion, file_path  # Return both emotion and file path
 
+def button_click(selected_audio):
+    emotion, file_path = predict_emotion(selected_audio)
+    return emotion, gradio.Interface.Play("rec/" + selected_audio)
+
 # Get the list of audio files for the dropdown
 audio_files_list = get_audio_files_list()
 
 # Loading Gradio interface
-dropdown = gr.Dropdown(label="Select Audio", choices=audio_files_list)
-button = gr.Button("Detect emotion")
+inputs = gr.Dropdown(label="Select Audio", choices=audio_files_list)
 outputs = [gr.outputs.Textbox(label="Predicted Emotion"), gr.outputs.Audio(label="Play Audio")]
 
-def button_click(selected_audio):
-    return predict_emotion(selected_audio)  # Call predict_emotion when button is clicked
+# Create the button
+sub_btn = gr.Interface.Button(label="Detect Emotion", elem_id="btn", onclick=button_click)
 
-title = "ML Speech Emotion Detection"
+title = "ML Speech Emotion Detection3"
 description = "Speechbrain powered wav2vec 2.0 pretrained model on IEMOCAP dataset using Gradio."
 
-# Create the Gradio interface
-interface = gr.Interface(
-    fn=button_click,  # Use the button_click function for the interface
-    inputs=[dropdown, button],
-    outputs=outputs,
-    title=title,
-    description=description
-)
-
+interface = gr.Interface(fn=predict_emotion, inputs=[inputs, sub_btn], outputs=outputs, title=title, description=description)
 interface.launch()