Update app.py

app.py

@@ -1,6 +1,4 @@
-# app.py
-
-# Import necessary libraries
+# Import Libraries
 import numpy as np
 import gradio as gr
 from sklearn.cluster import KMeans
@@ -24,7 +22,7 @@ caption_pipeline = pipeline(
 )
 
 # Translation Pipeline
-# Using facebook/mbart-large-50-many-to-many-mmt for higher-quality translations
+# Using facebook/mbart-large-50-many-to-many-mmt for translations
 # This model supports multiple languages and provides better translation quality for Arabic
 translation_pipeline = pipeline(
     "translation",
@@ -246,15 +244,16 @@ def translate_to_arabic(text):
         result = translation_pipeline(text)
         translated_text = result[0]['translation_text']
         
-        # Advanced Post-processing to remove repeated words
-        # This example uses a simple method; for more robust solutions, consider using NLP libraries
+        # Post-processing to remove repeated words
         words = translated_text.split()
+        seen = set()
         cleaned_words = []
         previous_word = ""
         for word in words:
             if word != previous_word:
                 cleaned_words.append(word)
-                previous_word = word
+                seen.add(word)
+            previous_word = word
         cleaned_translated_text = ' '.join(cleaned_words)
         
         return cleaned_translated_text