Create app.py

app.py

@@ -0,0 +1,57 @@
+import streamlit as st
+import PyPDF2
+import docx2txt
+from transformers import pipeline
+
+# Hugging Face translation pipeline
+translator_en = pipeline("translation", model="Helsinki-NLP/opus-mt-mul-en")  # Multilingual to English
+translator_ur = pipeline("translation", model="Helsinki-NLP/opus-mt-mul-ur")  # Multilingual to Urdu
+
+def extract_text_from_pdf(file):
+    """Extract text from a PDF file."""
+    pdf_reader = PyPDF2.PdfReader(file)
+    text = ""
+    for page in pdf_reader.pages:
+        text += page.extract_text()
+    return text
+
+def extract_text_from_word(file):
+    """Extract text from a Word file."""
+    return docx2txt.process(file)
+
+def translate_text(text, target_language):
+    """Translate text to the selected language."""
+    if target_language == "English":
+        return translator_en(text[:500]) if text else "No text found"
+    elif target_language == "Urdu":
+        return translator_ur(text[:500]) if text else "No text found"
+    return "Invalid translation choice."
+
+# Streamlit UI
+st.title("Multilingual Document Translator")
+st.write("Translate PDF or Word documents to English and Urdu quickly.")
+
+# File uploader
+uploaded_file = st.file_uploader("Upload a PDF or Word file", type=["pdf", "docx"])
+target_language = st.radio("Select the target language for translation", ["English", "Urdu"])
+
+if uploaded_file:
+    # Extract text
+    if uploaded_file.name.endswith(".pdf"):
+        text_content = extract_text_from_pdf(uploaded_file)
+    else:
+        text_content = extract_text_from_word(uploaded_file)
+
+    # Show extracted text preview
+    st.subheader("Extracted Text (Preview)")
+    st.write(text_content[:500] if text_content else "No content found in the file.")
+
+    # Perform translation
+    if st.button("Translate"):
+        if text_content:
+            st.subheader(f"Translated Text ({target_language})")
+            translation_results = translate_text(text_content, target_language)
+            translations = "\n".join([result['translation_text'] for result in translation_results])
+            st.text_area("Translation Output", translations, height=300)
+        else:
+            st.warning("No text found to translate. Please upload a valid document.")