import gradio as gr

import cv2
# Load the video file
cap = cv2.VideoCapture('video_file.mp4')
# Check if the video has audio
if cap.get(cv2.CAP_PROP_AUDIO_STATUS):
   # Read the video frames
   while True:
       ret, frame = cap.read()
       # Convert the frame to grayscale and apply thresholding
       gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
       _, thresh = cv2.threshold(gray, 100, 255, cv2.THRESH_BINARY)

       # Display the resulting image
       imshow("Thresholded Image", thresh)
       # Check if the user presses the 'q' key
       if cv2.waitKey(1) & 0xFF == ord('q'):
           # Break out of the loop
           break
else:
   print("No Audio Found")

# Release the video capture
cap.release()

# Load the audio file
audio, sr = librosa.load('audio_file.wav')

# Generate a new audio file with the same duration as the video
new_audio = np.zeros((len(frame), sr))
for i in range(len(frame)):
   # Calculate the time stamp for each pixel in the frame
   t = (i * framerate) + start_time
   
   # Add the corresponding value from the audio signal to the new audio array
   new_audio[i] = audio[int(t)]

# Save the new audio file
librosa.save('output_file.wav', new_audio, sr)