Update app.py

app.py

@@ -4,7 +4,6 @@ import torch
 import uvicorn
 import spacy
 import pdfplumber
-import moviepy.editor as mp
 import librosa
 import soundfile as sf
 import matplotlib.pyplot as plt
@@ -20,6 +19,7 @@ from pyngrok import ngrok
 from threading import Thread
 import time
 import uuid
+import subprocess  # Used for running ffmpeg commands
 
 # ✅ Ensure compatibility with Google Colab
 try:
@@ -260,12 +260,17 @@ def extract_text_from_pdf(pdf_file):
         raise HTTPException(status_code=400, detail=f"PDF extraction failed: {str(e)}")
 
 def process_video_to_text(video_file_path):
-    """Extract audio from video and convert to text."""
+    """Extract audio from video using ffmpeg and convert to text."""
     try:
         print(f"Processing video file at {video_file_path}")
         temp_audio_path = os.path.join("temp", "extracted_audio.wav")
-        video = mp.VideoFileClip(video_file_path)
-        video.audio.write_audiofile(temp_audio_path, codec='pcm_s16le')
+        # Use ffmpeg command to extract audio from the video file
+        cmd = [
+            "ffmpeg", "-i", video_file_path, "-vn",
+            "-acodec", "pcm_s16le", "-ar", "44100", "-ac", "2",
+            temp_audio_path, "-y"
+        ]
+        subprocess.run(cmd, check=True)
         print(f"Audio extracted to {temp_audio_path}")
         result = speech_to_text(temp_audio_path)
         transcript = result["text"]