Update app.py

app.py

@@ -1,22 +1,20 @@
-# Install required libraries
-# pip install -U openai-whisper gtts
+# Install dependencies if not already done in your environment
+# pip install transformers torch gtts ffmpeg-python
 
-import whisper
+from transformers import pipeline
 from gtts import gTTS
 import subprocess
 
-# Step 1: Extract Text from Video using Whisper
-def extract_text_from_video(video_path):
-    # Load the Whisper model
-    model = whisper.load_model("base")
+# Step 1: Extract Text from Audio using Hugging Face Transformers
+def extract_text_from_audio(audio_path):
+    # Load the ASR pipeline from Hugging Face with a Whisper-like model
+    transcriber = pipeline("automatic-speech-recognition", model="openai/whisper-base")
     
-    # Transcribe the video file's audio
-    result = model.transcribe(video_path)
+    # Transcribe the audio file
+    transcription = transcriber(audio_path)
+    text = transcription["text"]
     
-    # Extract the transcribed text
-    text = result["text"]
-    
-    # Save the transcribed text to a file (optional)
+    # Save transcribed text to a file (optional)
     with open("video_text.txt", "w") as f:
         f.write(text)
     
@@ -48,8 +46,8 @@ def add_voice_over_to_video(video_path, audio_path, output_video_path="output_vi
 
 # Run the complete process
 def main(video_path):
-    # Step 1: Extract text from video
-    text = extract_text_from_video(video_path)
+    # Step 1: Extract text from video/audio
+    text = extract_text_from_audio(video_path)
     print("Extracted Text:", text)
     
     # Step 2: Generate voice-over from extracted text