Update app.py

app.py

@@ -8,11 +8,19 @@ import numpy as np
 emotion_labels = {0: 'neutral', 1: 'calm', 2: 'happy', 3: 'sad', 4: 'angry', 5: 'fearful'}
 
 def process_video_audio(video_path, audio_path):
-    print(type(video_path))
-    print(type(audio_path))
+
+    print(video_path)
+
+    print(len(audio_path))
+
+    print(len(audio_path[0]))
+
+    audio_path = pt.tensor(audio_path[0])
+    
     train_visual = np.zeros([1, 120, 120, 3, 10])
     train_audio_wave = np.zeros([1, 261540])
-    train_audio_cnn = np.zeros([1, 150, 512, 1])
+    train_audio_cnn = np.zeros([1, 150, 512
+, 1])
     
     mfcc = torchaudio.transforms.MFCC(n_mfcc=150, melkwargs={"n_fft": 1022, "n_mels": 150})