Update app.py

app.py

@@ -4,7 +4,6 @@ import torch.cuda
 import whisper
 from whisper.tokenizer import LANGUAGES
 from vid_to_wav import extract_audio
-from face_emotion_detection import process_video
 gpu = torch.cuda.is_available()
 model = None
 
@@ -27,14 +26,13 @@ def analyze_transcription(text, duration):
 
 def transcribe(filepath, language, task):
     print(filepath)
-    video = process_video(filepath)
     audio, audio_file, duration = extract_audio(filepath)
     print(type)
     language = None if language == "Detect" else language
     text = model.transcribe(
         audio_file, task=task.lower(), language=language, fp16=gpu,
     )["text"].strip()
-    return video, text, analyze_transcription(text, duration)
+    return text, analyze_transcription(text, duration)
 
 
 def get_interface(model_name="medium"):
@@ -59,7 +57,7 @@ def get_interface(model_name="medium"):
                 info="Whether to perform X->X speech recognition or X->English translation",
             ),
         ],
-        outputs=[gr.Video(label="Emotion Analysis"),
+        outputs=[
                  gr.Textbox(label="Transcription", lines=26),
                  gr.Textbox(label="Speech Analysis", lines=4)],
         # theme=gr.themes.Default(),