Update app.py

app.py

@@ -1,6 +1,7 @@
 import gradio as gr
 import torch as pt
-import torchaudio
+import
+ torchaudio
 import cv2
 import os
 import numpy as np
@@ -63,10 +64,10 @@ def process_video_audio(video_path):
     train_audio_wave = tf.reshape(tf.convert_to_tensor(train_audio_wave.numpy(), dtype=tf.float16), (1, 20, 13077))
     train_audio_cnn = tf.convert_to_tensor(train_audio_cnn.numpy(), dtype=tf.float16)
 
-    return last_frame, train_visual, train_audio_wave, train_audio_cnn
+    return last_frame, audio_path, train_visual, train_audio_wave, train_audio_cnn
     
 def predict_emotion(video_path):
-    last_frame, train_visual, train_audio_wave, train_audio_cnn = process_video_audio(video_path)
+    last_frame, audio_path, train_visual, train_audio_wave, train_audio_cnn = process_video_audio(video_path)
     
     model = load_model("model_vui_ve.keras")
     
@@ -77,13 +78,13 @@ def predict_emotion(video_path):
     })
     
     predicted_label = np.argmax(predictions)
-    return last_frame, predicted_label
+    return last_frame, audio_path, predicted_label
 
 def predict_emotion_gradio(video_path):
     emotion_dict = {0: 'neutral', 1: 'calm', 2: 'happy', 3: 'sad', 4: 'angry', 5: 'fearful'}
-    last_frame, predicted_label = predict_emotion(video_path)
+    last_frame, audio_path, predicted_label = predict_emotion(video_path)
     predicted_emotion = emotion_dict[predicted_label]
-    return last_frame, predicted_emotion
+    return last_frame, audio_path, predicted_emotion
 
 iface = gr.Interface(
     fn=predict_emotion_gradio, 
@@ -92,6 +93,7 @@ iface = gr.Interface(
     ],
     outputs=[
         gr.Image(label="Last Frame"),
+        gr.Audio(label = "Audio")
         gr.Textbox(label="Predicted Emotion")
     ],
     title="Emotion Recognition from Video",