Update app.py

app.py

@@ -3,7 +3,7 @@ import gradio as gr
 import fitz  # PyMuPDF
 from langchain_community.vectorstores import FAISS
 from langchain_community.embeddings import HuggingFaceEmbeddings
-from langchain.text_splitter import RecursiveCharacterTextSplitter
+from langchain_text_splitters import RecursiveCharacterTextSplitter
 from langchain_google_genai import ChatGoogleGenerativeAI
 from langchain_core.prompts import ChatPromptTemplate
 from langchain_core.output_parsers import StrOutputParser
@@ -30,7 +30,7 @@ def create_vectorstore(text):
     return vectorstore
 
 # ----------------------------------------------------
-# 3️⃣ Initialize Google Gemini Model (through LangChain)
+# 3️⃣ Initialize Google Gemini Model
 # ----------------------------------------------------
 def get_model():
     os.environ["GOOGLE_API_KEY"] = os.getenv("GOOGLE_API_KEY", "")
@@ -43,16 +43,16 @@ def chat_with_pdf(pdf_file, user_input, history):
     if pdf_file is None:
         return history + [["❌ Please upload a PDF file first.", ""]]
 
-    # Step 1: Extract and create FAISS
+    # Extract and create FAISS
     pdf_text = extract_text_from_pdf(pdf_file)
     vectorstore = create_vectorstore(pdf_text)
     retriever = vectorstore.as_retriever(search_kwargs={"k": 3})
 
-    # Step 2: Retrieve relevant chunks
+    # Retrieve relevant chunks
     docs = retriever.get_relevant_documents(user_input)
     context = "\n\n".join([d.page_content for d in docs])
 
-    # Step 3: Generate Answer
+    # Generate Answer
     prompt = ChatPromptTemplate.from_template(
         """
         You are a helpful Urdu assistant. Answer in Urdu (Roman Urdu is fine if needed).
@@ -70,7 +70,7 @@ def chat_with_pdf(pdf_file, user_input, history):
     chain = prompt | llm | StrOutputParser()
     answer = chain.invoke({"context": context, "question": user_input})
 
-    # Optional: Urdu text-to-speech
+    # Text-to-speech in Urdu
     tts = gTTS(answer, lang="ur")
     tts.save("response.mp3")
 
@@ -78,7 +78,7 @@ def chat_with_pdf(pdf_file, user_input, history):
     return history, "response.mp3"
 
 # ----------------------------------------------------
-# 5️⃣ Gradio UI
+# 5️⃣ Gradio Interface
 # ----------------------------------------------------
 with gr.Blocks(title="📘 Urdu RAG Chatbot") as demo:
     gr.Markdown("## 🤖 Urdu RAG Chatbot — Ask questions from your PDF (Roman Urdu supported)")