Update app.py

app.py

@@ -22,7 +22,7 @@ def process_video_audio(video_path, audio_path):
         train_audio_wave[0, :] = wav[0][:261540]
     else:
         train_audio_wave[0, :len(wav[0])] = wav[0]
-    train_audio_cnn[0, :, :, 0] = mfcc(train_audio_wave[0]).numpy()
+    train_audio_cnn[0, :, :, 0] = mfcc(train_audio_wave[0])
     
     cap = cv2.VideoCapture(video_path)
     frame_idx = 0
@@ -43,16 +43,14 @@ def process_video_audio(video_path, audio_path):
             last_frame = frame
             frame_idx += 1
     cap.release()
-    
-
     predicted_emotion = "unknown"
 
-    return last_frame, predicted_emotion
+    return last_frame
 
 # Định nghĩa giao diện Gradio
 def gradio_interface(video, audio):
-    frame, emotion = process_video_audio(video, audio)
-    return gr.Image(frame, label="Processed Frame"), emotion
+    frame= process_video_audio(video, audio)
+    return frame
 
 iface = gr.Interface(
     fn=gradio_interface,
@@ -61,8 +59,7 @@ iface = gr.Interface(
         gr.Audio()
     ],
     outputs=[
-        gr.Image(),
-        gr.Textbox()
+        gr.Image()
     ],
     live=True,
     title="Video and Audio Processing with Emotion Recognition"