Update app.py

app.py

@@ -24,6 +24,7 @@ def process_video_audio(video_path):
     train_visual = pt.zeros([1, 120, 120, 3, 10])
     train_audio_wave = pt.zeros([1, 261540])
     train_audio_cnn = pt.zeros([1, 150, 512, 1])
+
     
     mfcc = torchaudio.transforms.MFCC(n_mfcc=150, melkwargs={"n_fft": 1022, "n_mels": 150})
     
@@ -70,7 +71,7 @@ def process_video_audio(video_path):
 def predict_emotion(video_path):
     last_frame, audio_path, train_visual, train_audio_wave, train_audio_cnn = process_video_audio(video_path)
     
-    model = load_model("model_vui_ve1024.keras")
+    model = load_model("model_vui_ve2392.keras")
     
     predictions = model.predict({
         "input_visual": train_visual,
@@ -82,7 +83,7 @@ def predict_emotion(video_path):
     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'}
+    emotion_dict = {0: 'neutral', 1: 'calm', 2: 'happy', 3: 'sad', 4: 'angry', 5: 'fearful', 6: 'disgust', 7: 'surprised'}
     last_frame, audio_path, predicted_label = predict_emotion(video_path)
     predicted_emotion = emotion_dict[predicted_label]
     return last_frame, audio_path, predicted_emotion