Update app.py

app.py

@@ -1,16 +1,14 @@
 import streamlit as st
 from moviepy.editor import VideoFileClip
 import whisper
-from translate import Translator
+from googletrans import Translator
 from gtts import gTTS
 import tempfile
 import os
+import numpy as np
 
 # Initialize Whisper model
-try:
-    whisper_model = whisper.load_model("base")  # Ensure the model is installed from the correct Whisper library
-except AttributeError:
-    st.error("Whisper model could not be loaded. Ensure that Whisper is installed from GitHub.")
+whisper_model = whisper.load_model("base")
 
 # Language options
 LANGUAGES = {
@@ -46,15 +44,22 @@ if video_file:
             os.remove(temp_video_path)
             st.stop()
 
-        # Transcribe audio using Whisper
+        # Transcribe audio using Whisper in chunks
         try:
-            result = whisper_model.transcribe(audio_path)
-            original_text = result["text"]
-            st.write("Original Transcription:", original_text)
+            # Load the audio file with Whisper
+            audio = whisper.load_audio(audio_path)
+            audio_segments = split_audio(audio, segment_length=30)  # Split into 30-second segments
+
+            original_text = ""
+            for segment in audio_segments:
+                result = whisper_model.transcribe(segment)
+                original_text += result["text"] + " "  # Concatenate transcriptions
+
+            st.write("Original Transcription:", original_text.strip())
 
             # Translate text to the target language
-            translator = Translator(to_lang=LANGUAGES[target_language])
-            translated_text = translator.translate(original_text)
+            translator = Translator()
+            translated_text = translator.translate(original_text.strip(), dest=LANGUAGES[target_language]).text
             st.write(f"Translated Text ({target_language}):", translated_text)
 
             # Convert translated text to speech
@@ -72,3 +77,13 @@ if video_file:
         os.remove(temp_video_path)
         os.remove(audio_path)
         os.remove(audio_output_path)
+
+def split_audio(audio, segment_length=30):
+    """Split audio into segments of specified length in seconds."""
+    total_length = audio.shape[1]  # Total length in seconds
+    segments = []
+    for start in range(0, total_length, segment_length):
+        end = min(start + segment_length, total_length)
+        segment = audio[:, start:end]  # Append the segment
+        segments.append(segment)
+    return segments